betaFeedback

WorksEngineering

System-Theoretic Analysis of Extended Cellular Potts Models for Tumor Invasion

Abstract

The Cellular Potts Model has emerged as a standard framework for modeling multicellular systems and metastatic progression. Yet, computational extensions incorporating cell movement within spatial gradients of Extracellular Matrix (ECM), an important element of the tissue microenvironment, lack theoretical foundations. This work provides a system-theoretic formulation and analysis of Extended Cellular Potts Models (ECPMs) that explicitly couple tumor dynamics with ECM degradation. We reformulate the ECPM as an interconnected dynamical system: a stochastic process describing a network of locally interacting tumor cells coupled to a jump-linear system describing ECM evolution. By exploiting the sparse energy structure of the model, we characterize asymptotic ECM behavior, transition probabilities with uniform bounds, and global system convergence. In one-dimensional settings, we derive closed-form expressions for invasion time probabilities, revealing how mean invasion time depends on local ECM gradients, number of cells and their tendency to maintain a target volume. Numerical simulations validate theoretical predictions, establishing a principled foundation for analyzing haptotaxis-driven invasion in Potts-based models.

Introduction

Understanding the complex mechanisms governing tumor invasion and progression is a crucial goal in cancer research. Mathematical modeling plays a pivotal role in replicating the underlying principles of tumor dynamics, offering a structured framework to test hypotheses and interpret experimental findings. One class of models, the Cellular Potts Model (CPM), has emerged as a versatile tool to describe multicellular systems, capturing spatially resolved cell behaviors and interactions across multiple scales. A tumor can be viewed as a complex system with local interactions, in which agents (cells) interact locally with their neighbors and with a shared environment, giving rise to emergent collective behaviors such as invasion. Traditionally, CPMs depict tissues as ensembles of cells governed by energy-based rules that account for cell adhesion and volume constraints. This approach allows the emergent behavior of cellular assemblies to be studied as a consequence of local, cell-cell interactions.

However, real tumor environments are far more complex, shaped not only by intercellular adhesion but also by interactions between cells and their surrounding extracellular matrix (ECM). These interactions give rise to a key process in tumor invasion where cells move along gradients of ECM concentration or adhesion strength. Classical CPMs account only for the intercellular interaction network, and neglect the feedback between the cells and the evolving environment. Building on the foundational analysis provided in [1], which discusses the theoretical properties of standard CPMs, our work focuses on a specific subclass of Extended Cellular Potts Models (ECPMs) that explicitly incorporates cell-ECM interactions to model haptotaxis [2]. Within this framework, we extend classical CPM theory by leveraging system-theoretic tools to address more realistic tumor invasion models, thus offering new insights into the mechanisms by which cell-ECM coupling influences invasion dynamics.

Literature Review: In the following, we present the literature review divided into three categories: (i) early theoretical studies on CPMs, (ii) applied CPMs in cancer biology and their extensions ECPMs, and (iii) related works in the systems-and-control community.

Early work [3] introduces CPMs to simulate cell sorting via differential adhesion and energy minimization principles. Subsequent mathematical studies analyze critical behavior, phase transitions, and Gibbs measures to establish equilibrium properties in Potts models [4], [5], [6]. Rigorous analytical frameworks further characterize CPM dynamics—including ergodicity, convergence, and parameter sensitivity [1]—and connected discrete stochastic rules to continuum PDE limits [7], [8]. Despite these advances, most theoretical analyses have focused on the statistical mechanics or continuum limits of CPMs without explicitly incorporating ECM dynamics or cell–environment feedback, which are central to haptotaxis-driven invasion.

In biological applications CPMs have successfully modeled morphogenesis [9], cancer invasion driven by intercellular adhesion [10], and drug treatment failure dynamics [11]. Recent extensions, i.e. ECPMs for tumor invasion, explicitly incorporate ECM interactions, modeling tumor invasion [12], angiogenesis driven by matrix gradients [13], dynamic ECM remodeling in wound healing [14], and broader tissue organization [15], [16]. While these works provide powerful computational tools, they largely remain simulation-oriented, with limited theoretical formalization and results on the underlying dynamics. In particular, a rigorous theoretical understanding of tumor-cells-ECM feedback within ECPMs is still missing.

From a systems-and-control perspective, several recent works have addressed related classes of stochastic and interconnected dynamical systems. Markov jump linear systems have received renewed attention in the analysis and design of stochastic control systems [17], and switched and positive linear system formulations have been employed to study analogous transition mechanisms over networks [18]. In parallel, a substantial body of work in the network-control community has focused on discrete-time Markov-chain models of spreading and diffusion processes over networks, which, like the ECPM, are stochastic dynamics driven by local interactions. These works provided stability, convergence, and estimation results for epidemic and related processes [19], [20], [21], [22]. Control-theoretic tools have also been applied directly to cellular automata to study morphogenetic processes [23], and, more recently, to the coordination and control of biological cell populations, ranging from microbial consortia to spatially organized multi-agent systems [24], [25], [26]. Bio-inspired swarm models have similarly been proposed to describe cancer progression in control-theoretic terms [27], while computational approaches from the systems-biology community have investigated how ECM manipulation affects local tumor invasion [28]. These contributions reflect a growing interest in applying systems theory to biologically motivated stochastic and networked models, a direction that directly inspires our approach.

Contributions: The contributions of this paper are as follows. We introduce a novel system-theoretic framework for analyzing ECPMs that explicitly incorporate cell-ECM interactions responsible for haptotaxis-driven tumor invasion. Unlike classical CPMs, typically explored through empirical simulations, we establish a rigorous analytical foundation for these models by reformulating the ECPM as a closed-loop dynamical system composed of two interconnected subsystems: (i) a stochastic process governing the tumor cell configuration and (ii) a jump-linear system describing ECM evolution. We derive analytical results characterizing key properties of the coupled system, including the asymptotic behavior of the ECM field, uniform bounds on transition probabilities of tumor configurations, and convergence properties of the global dynamics. These results provide the first rigorous theoretical description of haptotaxis within the CPM framework. Moreover, in one-dimensional settings we exploit the sparse energy structure of the model to obtain closed-form expressions for invasion time probabilities. These expressions reveal how invasion time and stochastic propagation depend explicitly on haptotactic strength, ECM degradation rates, number of cells and their target volumes. Numerical simulations validate the theoretical predictions, confirming the analytical results for estimation of mean invasion time. This work bridges computational and theoretical analysis of ECPMs, by establishing a principled foundation for analyzing ECM-driven tumor invasion in Potts-based models. Finally, the theoretical tools for interconnected stochastic and jump-linear systems may pave the way to a broader class of control systems inspired by biology.

Paper Organization: The paper is organized as follows. Section II introduces the standard ECPM formalism, covering lattice structure, energy formulation, and the conventional algorithm used to simulate these models. Section III reformulates the ECPM as an interconnected dynamical system coupling stochastic cell configuration dynamics with jump-linear ECM evolution, emphasizing the sparse energy structure. Section IV analyzes system properties: asymptotic ECM behavior, transition probabilities and their bounds, and global convergence to absorbing states. Section V derives closed-form invasion time probability expressions for one-dimensional ECPMs. Section VI validates theoretical predictions through Monte Carlo simulations.

Notation: For any finite set SS, ∣S∣|S| denotes its cardinality. We use ∥⋅∥\|\cdot\| for the Euclidean norm. Any function F:S→RF:S\to\mathbb{R} is identified, up to isomorphism, with a vector in R∣S∣\mathbb{R}^{|S|} and is denoted by the same symbol. We write δ(⋅,⋅)\delta(\cdot,\cdot) for the Kronecker delta, P(⋅)P(\cdot) for probability, E[⋅]\mathbb{E}[\cdot] for expectation, and U(S)\mathcal{U}(S) for the uniform distribution on SS. Throughout the paper we will use the terms “grid” and “lattice” interchangeably, with “lattice” emphasizing the mathematical structure.

Conventional Algorithmic Formulation of Extended Cellular Potts Models

In this section, we recall the standard ECPM algorithmic formulation for simulating ECM-driven tumor invasion and set up the notation for the system-theoretic formalization and analysis developed in the next sections.

In ECPMs, the spatial domain is discretized into a regular grid of sites, each representing a small patch of ECM that can be occupied by cells or remain empty. Specifically, let us consider a two-dimensional finite lattice G:=X×Y⊂Z2G:=\mathcal{X}\times\mathcal{Y}\subset\mathbb{Z}^{2} with dimensions nx:=∣X∣n_x:=|\mathcal{X}| and ny:=∣Y∣n_y:=|\mathcal{Y}|. On GG there exist boundary conditions consisting of zero-flux at the top and bottom boundaries, and periodic conditions along the left and right sides of the grid. Zero-flux at the top/bottom models confinement within the epithelial boundary layer of a lumen, preventing motion in that direction. Instead, the periodic lateral boundary conditions approximate a small portion of a much larger domain. Let us introduce a finite set of labels L⊂NL\subset\mathbb{N}. Then, each grid point p=(xp,yp)∈Gp=(x_p,y_p)\in G is associated with a label through the map

σ:G→Lp↦σp,(1)\begin{aligned} \sigma&: G \to L\\ p &\mapsto\sigma_p, \tag*{(1)} \end{aligned}

Here, each label σp∈L\sigma_p\in L identifies the entity occupying the grid point p∈Gp\in G, which can belong to a tumor cell or empty space. Thus, the label map in (1) provides a complete representation of the spatial configuration of all entities on the grid. Moreover, each label ℓ∈L\ell\in L is associated with a specific type via the following mapping

τ:L→Iℓ↦τℓ,(2)\begin{aligned} \tau&: L \to\mathbb{I}\\ \ell&\mapsto\tau_\ell, \tag*{(2)} \end{aligned}

where Γ⊂N\Gamma\subset\mathbb{N} is a finite set containing grid point types that represent either cell types (e.g., tumor cells or normal cells) or the ECM. We now introduce the ECM density scalar field ff over the discrete lattice. Formally,

f:G→R≥0p↦fp,(3)\begin{aligned} f &: G \to\mathbb{R}_{\ge0}\\ p &\mapsto f_p, \tag*{(3)} \end{aligned}

that evolves over time due to degradation by proteolytic enzymes secreted by tumor cells. In practice, the ECM field is uniformly initialized on the lattice to a constant value finit∈R>0f_{\mathrm{init}} \in\mathbb{R}_{>0}, i.e. fp=finitf_p = f_{\mathrm{init}} for all p∈Gp \in G. Before presenting the specific

update rule of (3), we introduce some quantities defined on the grid. Consider two arbitrary grid points p,q∈Gp, q \in G, then we equip the grid with the following Chebyshev distance metric

dG(p,q):=max⁡(Δx(p,q),∥yp−yq∥),d_G(p,q) := \max\left(\Delta_x(p,q), \lVert y_p-y_q\rVert\right),

where Δx(p,q):=min⁡(∥xp−xq∥,nx−∥xp−xq∥)\Delta_x(p,q) := \min\left(\lVert x_p-x_q\rVert,n_x-\lVert x_p-x_q\rVert\right) accounts for the periodic boundary conditions along the horizontal direction. The metric dGd_G is the shortest-path distance of the undirected graph (G,E)(G,\mathcal{E}), with edges E:={(p,q)∈G2:dG(p,q)=1}\mathcal{E} := \{(p,q) \in G^2 : d_G(p,q)=1\}, whose nodes are the lattice sites. Then, the closed ball of radius d∈Nd \in\mathbb{N} centered at grid point pp can be expressed as B‾(p,d):={q∈G:dG(p,q)≤d}\overline{B}(p,d) := \{q \in G : d_G(p,q) \le d\}. For γ∈Γ\gamma\in\Gamma and d>0d > 0 we define the number of cells of type γ\gamma within distance dd from pp by

nd(p,γ,σ):=∣{ℓ∈σ(B‾(p,d)):τℓ=γ}∣.n_d(p,\gamma,\sigma) := \left|\left\{\ell\in\sigma\left(\overline{B}(p,d)\right) : \tau_\ell=\gamma\right\}\right|.

Next, we introduce the model parameters that will appear in the ff dynamics. For each cell type γ∈Γ\gamma\in\Gamma, let cγz≥0c^z_\gamma\ge0 be the enzyme secretion rate, aγ≥0a_\gamma\ge0 the spatial decay coefficient, and Dγ∈ND_\gamma\in\mathbb{N} the maximal influence radius. Let cγf≥0c^f_\gamma\ge0 denote the ECM degradation rate induced by cells of type γ\gamma. With these definitions in place, the proteolytic enzyme produced at p∈Gp \in G by cells of type γ\gamma is given by

z(p,γ,σ):=cγz∑d=1Dγnd(p,γ,σ)e−aγd,(4)z(p,\gamma,\sigma) := c^z_\gamma\sum_{d=1}^{D_\gamma} n_d(p,\gamma,\sigma)e^{-a_\gamma d}, \tag*{(4)}

from which follows the ECM update rule, expressed as

fpk+1=(1−∑γ∈Γcγfz(p,γ,σk))fpk,∀p∈G,(5)f_p^{k+1}=\left(1-\sum_{\gamma\in\Gamma}c^f_\gamma z(p,\gamma,\sigma^k)\right)f_p^k,\quad\forall p\in G, \tag*{(5)}

where fpkf_p^k and fpk+1f_p^{k+1} denote the ECM density at lattice point pp at two consecutive iterations. Note that the ECM update at each node depends only on the labels within graph distance DγD_\gamma, i.e., degradation is driven by the local neighborhood of each node. To ensure physical consistency, we assume that the parameters cγzc^z_\gamma and cγfc^f_\gamma are chosen such that

∑γ∈Γcγfz(p,γ,σ)≤1∀p∈G,(6)\sum_{\gamma\in\Gamma}c^f_\gamma z(p,\gamma,\sigma)\le1\quad\forall p\in G, \tag*{(6)}

namely, the ECM density cannot become negative after degradation. Next, we define the interaction energy between neighboring grid points via the map J:Γ×Γ→R≥0J:\Gamma\times\Gamma\to\mathbb{R}_{\ge0}, which quantifies the interfacial energy between pairs of cell or medium types. Biologically, JJ captures the energetic cost of maintaining contact between two entities, reflecting effects such as cell-cell or cell–matrix adhesion and cortical tension. In this work, we adopt the standard ECPM convention (see [10, 11]), where the interaction energy is specified as

J(γECM,γECM)=0(7)J(\gamma_{\mathrm{ECM}},\gamma_{\mathrm{ECM}})=0 \tag*{(7)}
J(γc,γc)=Jc,c(8)J(\gamma_c,\gamma_c)=J_{c,c} \tag*{(8)}
J(γc,γECM)=J(γECM,γc)=JECM,c,(9)J(\gamma_c,\gamma_{\mathrm{ECM}})=J(\gamma_{\mathrm{ECM}},\gamma_c)=J_{\mathrm{ECM},c}, \tag*{(9)}

where Jc,c∈R≥0J_{c,c} \in\mathbb{R}_{\ge0} and JECM,c∈R≥0J_{\mathrm{ECM},c} \in\mathbb{R}_{\ge0} denote the cell-cell and cell-ECM adhesion energies, respectively. Here, γECM∈Γ\gamma_{\mathrm{ECM}} \in\Gamma is the type corresponding to an empty grid point covered by ECM, and γc∈Γ\gamma_c \in\Gamma is the tumor cell type. This reflects the assumption that ECM-ECM interfaces are energetically neutral, while cell-cell and cell-ECM contacts are penalized according to their adhesion strengths. Consequently, the interaction energy between labels is defined as

J~:L×L→R≥0(ℓ1,ℓ2)↦J(τℓ1,τℓ2)[1−δ(ℓ1,ℓ2)],(10)\begin{aligned} \widetilde{J}: L \times L &\to\mathbb{R}_{\ge0}\\ (\ell_1,\ell_2) &\mapsto J(\tau_{\ell_1},\tau_{\ell_2})\left[1-\delta(\ell_1,\ell_2)\right], \tag*{(10)} \end{aligned}

where the Kronecker delta ensures that only interfaces between different labels contribute to the energy. Figure 1 illustrates these cell-surface interactions as regulated by J~\widetilde{J}. Define the set of neighbors of a point pp as Np:={q∈G:(p,q)∈E}N_p := \{q \in G : (p,q) \in E\}. Hence, a cell is a set of nodes sharing the same label, and adhesion energy arises only along the edges connecting nodes with different labels. Then, we introduce the core term in CPMs, i.e. the total energy of the system H:L∣G∣→R≥0H : L^{|G|} \to\mathbb{R}_{\ge0}, represented by the Hamiltonian function H(σ):=HJ(σ)+HV(σ)H(\sigma) := H_J(\sigma) + H_V(\sigma), where

Figure 1. Cell-surface interactions in the ECPM are regulated by energy coefficients Jc,cJ_{c,c} (cell-cell) and JECM,cJ_{\mathrm{ECM},c} (cell-ECM) adhesion. Here, a single tumor cell (red) interacts with neighboring cells (green and blue) and ECM (gray).

HJ(σ):=12∑p∈G∑q∈NpJ~(σp,σq)(11)H_J(\sigma):=\frac{1}{2}\sum_{p\in G}\sum_{q\in N_p}\widetilde{J}(\sigma_p,\sigma_q) \tag*{(11)}
HV(σ):=λ∑ℓ∈L∖{ℓ0}(∑w∈Gδ(σw,ℓ)−VT)2.(12)H_V(\sigma):=\lambda\sum_{\ell\in L\setminus\{\ell_0\}}\left(\sum_{w\in G}\delta(\sigma_w,\ell)-V_T\right)^2. \tag*{(12)}

Here, HJH_J collects all local interaction contributions between neighboring sites, while HVH_V is the volume penalty term that favors label-wise occupancy close to a target value. The parameter λ∈R>0\lambda\in\mathbb{R}_{>0} weighs the strength of the volume constraint, thus regulating how strongly cells tend to maintain their target volume. In (10), ℓ0\ell_0 represents the label corresponding to empty lattice sites covered by ECM. The label ℓ0\ell_0 is excluded from the sum since empty/ECM sites do not contribute to HH.

ECPMs (and CPMs) are simulated in practice through a Markov chain Monte Carlo method, the so-called Modified Metropolis Algorithm (MMA) [29, 30]. The latter constitutes an adaptation of the classical Metropolis algorithm [31] in the context of Potts models. The MMA is described step-by-step in Algorithm 1. In equations (11) and (12), KECM≥0K_{\mathrm{ECM}} \ge0 is a model parameter.

Algorithm 1 MMA for simulating ECPM

  1. Start from an initial configuration (σ,f)(\sigma,f)

  2. repeat

  3. Select grid point p∼U(G)p \sim\mathcal{U}(G)

  4. Pick nearest neighbor q∼U(Np)q \sim\mathcal{U}(N_p)

  5. if σp=σq\sigma_p=\sigma_q then

  6.   goto step 3

  7. end if

  8. Define trial configuration σ′\sigma' such that

σz′=σz+(σp−σq)δ(z,q)∀z∈G\sigma'_z=\sigma_z+(\sigma_p-\sigma_q)\delta(z,q)\quad\forall z\in G
  1. Evaluate energy change ΔH=H(σ′)−H(σ)\Delta H=H(\sigma')-H(\sigma)

  2. Compute total energy

ΔE(σ,f)=ΔH+KECM(fp−fq)(13)\Delta E(\sigma,f)=\Delta H+K_{\mathrm{ECM}}(f_p-f_q) \tag*{(13)}
  1. if ΔE(σ,f)<0\Delta E(\sigma,f)<0 then

  2. Accept the copy and update σq←σp\sigma_q\leftarrow\sigma_p

  3. else

  4. Accept the change σq←σp\sigma_q\leftarrow\sigma_p with probability

e−ΔE(σ,f)β(14)e^{-\frac{\Delta E(\sigma,f)}{\beta}} \tag*{(14)}
  1. end if

  2. Update fz ∀z∈Gf_z\ \forall z\in G via dynamics (5)

  3. until stopping criterion

that weighs the influence of ECM gradients on cell migration, while parameter β>0\beta> 0 influences the stochastic behavior of the dynamics, in fact it is usually related to environmental noise. Note that each MMA iteration is a local asynchronous update, depending only on the neighborhood of the selected node. While the algorithmic description enables simulation of

ECPM dynamics, it does not provide analytical insight into the underlying system properties. As an example, Figure 2 shows a simulation on a 2D lattice, where varying the adhesion energies Jc,cJ_{c,c} and JECM,cJ_{ECM,c} produces different invasion patterns. However, such simulations are descriptive and do not support a systematic analysis of the model’s behavior. To overcome this limitation, the next section reformulates the ECPM in a system-theoretic framework, enabling a principled study of its dynamics.

ECPM simulation on a 2D lattice showing tumor invasion through ECM at different adhesion energies

Figure 2. ECPM simulation on a 2D lattice showing tumor invasion (red) through ECM (gray) at different adhesion energies Jc,cJ_{c,c} (cell-cell) and JECM,cJ_{ECM,c} (cell-ECM).

ECPM as Sparse Interconnected Systems

In this section, rather than treating the ECPM as an algorithmic rule, we regard it as an interconnected dynamical system composed of two distinct yet coupled subsystems: (i) a stochastic process describing a network of locally interacting tumor cells, influenced by the ECM substrate, and (ii) a jump-linear system governing the ECM density field, whose transition matrix changes due to the current tumor mass configuration. We start by introducing the ECM transition matrix underlying dynamics (5) as follows

A(σ):=diag⁡p∈G{Ap(σ)},(15)A(\sigma) := \operatorname{diag}_{p \in G}\{A_p(\sigma)\}, \tag*{(15)}

with the local terms defined as

Ap(σ):=1−αp(σ)=1−∑γ∈Γcγpz(p,γ,σ),(16)A_p(\sigma) := 1-\alpha_p(\sigma)=1-\sum_{\gamma\in\Gamma} c_\gamma^p z(p,\gamma,\sigma), \tag*{(16)}

where αp(σ)\alpha_p(\sigma) represents the total degradation rate at site pp induced by all cell types present at current configuration σ\sigma. Here, Ap(σ)∈[0,1]A_p(\sigma) \in[0,1] since 0≤αp(σ)≤10 \leq\alpha_p(\sigma) \leq1 due to condition (6) and being the parameters within (4) non-negative. In view of the definition in (5), the ECM field can be identified with a vector f∈R≥0∣G∣f \in\mathbb{R}_{\geq0}^{|G|} (see Notation). Accordingly, its dynamics can be rewritten as the following jump-linear system

fk+1=A(σk)fk,(17)f^{k+1} = A(\sigma^k)f^k, \tag*{(17)}

with σk\sigma^k and fkf^k being the cell mass configuration and ECM field on the lattice at discrete-time (iteration) k∈Nk \in\mathbb{N}, respectively. Moreover, since the acceptance probability in (12) depends on the local ECM values fpf_p and fqf_q (see ΔE\Delta E definition at MMA step 11), the σ\sigma-dynamics are thus influenced by ff. Formally, the stochastic process associated with the evolution of σ\sigma can be generally written as

σk+1∼P(⋅∣σk,fk),(18)\sigma^{k+1} \sim P(\mathord{\cdot}\mid\sigma^k,f^k), \tag*{(18)}

where the transition probability is given by

P(⋅∣σ,f):={1if ΔE(σ,f)≤0,e−ΔE(σ,f)βotherwise.(19)P(\mathord{\cdot}\mid\sigma,f) := \begin{cases} 1 & \text{if } \Delta E(\sigma,f) \leq0,\\ e^{-\frac{\Delta E(\sigma,f)}{\beta}} & \text{otherwise}. \end{cases} \tag*{(19)}

that is a piecewise Boltzmann-weighted probability derived from the MMA steps 11-14 (see Algorithm 1). The resulting feedback interconnection between the subsystems (15) and (16) is pictured in Figure 3. Based on this reformulation, we analyze the Hamiltonian variation ΔH\Delta H. Local cell-cell and cell-ECM interactions produce a sparse energy structure determined by the interaction graph (G,E)(G,\mathcal{E}) as per (9). This sparsity links the stochastic process (16) of interacting tumor cells directly to local energy terms, enhancing analytical tractability. This property underpins the derivation of closed-form expressions for invasion time probabilities presented in § 5. We formalize this sparsity property in the following lemma.

Figure 3. System-theoretic view of the ECPM. The discrete stochastic process is coupled to a jump-linear system.

Lemma III.1 (Sparse Energy Decomposition). Consider an ECPM at current state (σ,f)(\sigma,f). Let σ′\sigma' be an adjacent configuration that differs from σ\sigma only at neighboring lattice sites p∈Gp \in G and q∈Npq \in\mathcal{N}_p. Then, the total energy change ΔE(σ,f)\Delta E(\sigma,f) can be decomposed into sparse energy terms as follows

ΔE(σ,f)=ΔHJ(σ)+ΔHV(σ)+KECM(fp−fq),\Delta E(\sigma,f) = \Delta H_J(\sigma) + \Delta H_V(\sigma) + K_{\mathrm{ECM}}(f_p-f_q),

where the adhesion-energy change is given by

ΔHJ(σ)=∑w∈NqJ~(σp,σw)−J~(σq,σw),(20)\Delta H_J(\sigma) = \sum_{w \in\mathcal{N}_q} \widetilde{J}(\sigma_p,\sigma_w) - \widetilde{J}(\sigma_q,\sigma_w), \tag*{(20)}

and the volume-energy change simplifies to

ΔHV(σ)=λ∑ℓ∈{σp,σq}[1+2ρ(σp,ℓ)(Vℓ−VT)](1−δ(ℓ0,ℓ)),(21)\Delta H_V(\sigma) = \lambda\sum_{\ell\in\{\sigma_p,\sigma_q\}} \left[1 + 2\rho(\sigma_p,\ell)(V_\ell-V_T)\right]\left(1-\delta(\ell_0,\ell)\right), \tag*{(21)}

with

Vℓ:=∑w∈Gδ(σw,ℓ),ρ(ℓ1,ℓ2):=2δ(ℓ1,ℓ2)−1∈{−1,1}.V_\ell:= \sum_{w \in G}\delta(\sigma_w,\ell), \qquad\rho(\ell_1,\ell_2) := 2\delta(\ell_1,\ell_2)-1 \in\{-1,1\}.

Proof. See Appendix A.

This result highlights how the total energy change decomposes into three sparse terms that determine the stochastic evolution in (16) following the probability rule in (17). Therefore, the state dynamics of the ECPM can be interpreted as a stochastic process driven by a sparse energy functional, coupled to a jump-linear system representing the ECM degradation. This novel interpretation clarifies the feedback mechanisms between tumor cells and the ECM, making the interaction dynamics in this class of models more explicit and analytically tractable.

System Analysis and Properties

In this section we analyze the dynamical properties of the ECPM as formulated in Section III. We begin by presenting a result on the asymptotic behavior of the ECM field ff, and then we characterize the transition probabilities related to the σ\sigma-dynamics. Then, we define uniform lower and upper bounds on the transition probabilities as a function of model parameters. Finally, we fully characterize the global system behavior and its convergence properties. In the following, we denote by Sk=(σk,fk)\mathcal{S}^k = (\sigma^k,f^k) the global ECPM state at iteration kk, with dynamics as in (15)-(16) (see also Figure 3).

Asymptotic Behavior of ECM Dynamics

We first establish that the ECM concentration ff remains bounded and converges to a steady-state value, regardless of the evolution of the cell mass configuration σ\sigma. We additionally characterize the limiting value of ff when the cell mass configuration is held fixed. Formally, we have the following property.

Lemma IV.1 (Asymptotic Behavior of ECM). If {Sk}k∈N\{S^k\}_{k\in\mathbb{N}} is a trajectory for the interconnection of (15)–(16), then lim⁡k→∞fk=fˉ∈[0,finit]∣G∣\lim_{k\to\infty} f^k = \bar{f} \in[0,f_{\mathrm{init}}]^{|G|}. In particular, if the cell mass configuration is fixed, i.e. σk=σˉ\sigma^k=\bar{\sigma} for all k∈Nk\in\mathbb{N}, then

lim⁡k→∞fpk={0if Ap(σˉ)∈[0,1),fp0if Ap(σˉ)=1,∀p∈G.\lim_{k\to\infty} f_p^k = \begin{cases} 0 & \text{if } A_p(\bar{\sigma})\in[0,1),\\ f_p^0 & \text{if } A_p(\bar{\sigma})=1 \end{cases} ,\qquad\forall p\in G.

Proof. See Appendix B.

Transition Probabilities of Cell Mass Configuration

Let us start by introducing the difference set between two configurations as Θ(σ1,σ2):={p∈G:σ1,p≠σ2,p}\Theta(\sigma_1,\sigma_2):=\{p\in G:\sigma_{1,p}\ne\sigma_{2,p}\}. Then, the distance function dL(σ1,σ2)=∣Θ(σ1,σ2)∣d_L(\sigma_1,\sigma_2)=|\Theta(\sigma_1,\sigma_2)| counts the number of different points between two configuration. We state that two distinct configurations σ1\sigma_1 and σ2\sigma_2 are adjacent if and only if dL(σ1,σ2)=1d_L(\sigma_1,\sigma_2)=1. Given the current ECPM state SkS^k, the transition probability P(σk+1=σ′∣Sk)P(\sigma^{k+1}=\sigma'\mid S^k) for the system to evolve to the configuration σ′\sigma' is zero either if the two configurations are not adjacent, or the two configurations are adjacent with Θ(σk,σ′)={p}\Theta(\sigma^k,\sigma')=\{p\}, but they differ by an outlier (an isolated label), i.e. σp′≠σq′\sigma'_p\ne\sigma'_q, ∀q∈Np\forall q\in\mathcal{N}_p. Otherwise, the probability of moving to σ′\sigma' is always nonzero. We now characterize these transition probabilities. To this end, we first introduce some notation. For a given configuration σ\sigma, let NU(σ)≤∣E∣N_U(\sigma)\leq|\mathcal{E}| denote the number of edges in E\mathcal{E} with the same label, i.e., NU(σ):=∣{(p,q)∈E:σp=σq}∣N_U(\sigma):=|\{(p,q)\in\mathcal{E}:\sigma_p=\sigma_q\}|. Next, we partition the set of configurations adjacent to σ\sigma based on the sign of the energy variation. Define Σa+(S)\Sigma_a^+(\mathcal{S}) as the set of configurations σ′\sigma' adjacent to σ\sigma for which the energy change is positive, specifically

Σa+(S):={σ′∈L∣G∣:Θ(σ,σ′)={p}, ΔE(σ′∣S)>0}.\Sigma_a^+(\mathcal{S}):=\left\{\sigma'\in\mathcal{L}^{|G|}:\Theta(\sigma,\sigma')=\{p\},\ \Delta E(\sigma'|\mathcal{S})>0\right\}.

Similarly, define Σa−(S)\Sigma_a^-(\mathcal{S}) as the set of configurations σ′\sigma' adjacent to σ\sigma for which the energy change is non-positive, i.e.

Σa−(S):={σ′∈L∣G∣:Θ(σ,σ′)={p}, ΔE(σ′∣S)≤0}.\Sigma_a^-(\mathcal{S}):=\left\{\sigma'\in\mathcal{L}^{|G|}:\Theta(\sigma,\sigma')=\{p\},\ \Delta E(\sigma'|\mathcal{S})\leq0\right\}.

Then, all transient configurations adjacent to σ\sigma belong to the set Σa(S):=Σa+(S)∪Σa−(S)\Sigma_a(\mathcal{S}):=\Sigma_a^+(\mathcal{S})\cup\Sigma_a^-(\mathcal{S}). With these definitions in place, we can now express the transition probabilities as follows. For a configuration σ′∈Σa−(S)\sigma'\in\Sigma_a^-(\mathcal{S}) (favorable energy change), the transition probability is

P(σk+1=σ′∣Sk)=1∣E∣−NU(σk),(22)P(\sigma^{k+1}=\sigma'|S^k)=\frac{1}{|\mathcal{E}|-N_U(\sigma^k)}, \tag*{(22)}

where the denominator reflects that only adjacent configurations with different labels can be reached in one step, excluding the NU(σ)N_U(\sigma) edges with the same label. Instead, for a configuration σ′∈Σa+(S)\sigma'\in\Sigma_a^+(\mathcal{S}) (unfavorable energy change), the transition probability is

P(σk+1=σ′∣Sk)=exp⁡(−ΔE(Sk)/β)∣E∣−NU(σk).(23)P(\sigma^{k+1}=\sigma'|S^k)=\frac{\exp\left(-\Delta E(S^k)/\beta\right)}{|\mathcal{E}|-N_U(\sigma^k)}. \tag*{(23)}

Finally, the probability of remaining in the same σ\sigma is

P(σk+1=σk∣Sk)=∑σ′∈Σa+(Sk)1−exp⁡(−ΔE(Sk)/β)∣E∣−NU(σk),(24)P(\sigma^{k+1}=\sigma^k|S^k)=\sum_{\sigma'\in\Sigma_a^+(S^k)}\frac{1-\exp\left(-\Delta E(S^k)/\beta\right)}{|\mathcal{E}|-N_U(\sigma^k)}, \tag*{(24)}

which accounts for σ\sigma-transitions to energetically unfavorable configurations that do not occur. Based on the above expressions, we can now provide explicit uniform lower and upper bounds for the transition probabilities in terms of the model parameters and grid dimensions. To state the bound, let ΔHmax⁡,ΔHmin⁡∈R\Delta H_{\max},\Delta H_{\min}\in\mathbb{R} be constants such that, for every feasible adjacent configuration σ′∈Σa(S)\sigma'\in\Sigma_a(\mathcal{S}) it holds that ΔHmin⁡≤H(σ′)−H(σ)≤ΔHmax⁡\Delta H_{\min}\leq H(\sigma')-H(\sigma)\leq\Delta H_{\max}.

Lemma IV.2 (Bounds on Transition Probabilities). For any trial configuration σ′∈Σa(Sk)\sigma'\in\Sigma_a(S^k), the corresponding transition probability admits the uniform lower and upper bounds

ρ≤P(σk+1=σ′∣Sk)≤12whereρ:=e−(ΔHmax⁡+KECMfinit)/βnx(8ny−6).\rho\leq P(\sigma^{k+1}=\sigma'|S^k)\leq\frac{1}{2} \qquad\text{where}\qquad \rho:=\frac{e^{-(\Delta H_{\max}+K_{ECM}f_{\mathrm{init}})/\beta}}{n_x(8n_y-6)}.

Proof. See Appendix C.

Interconnected System Evolution and Absorbing States

We finally present the main result of this section, namely the characterization of the interconnected system dynamics formulated in Section III. This result leverages the properties established in previous lemmas to describe the global system behavior. Considering the coupled dynamics in (15)-(16), the state space is Σ×F\Sigma\times F, where Σ⊂L∣G∣\Sigma\subset L^{|G|} represents the finite set of label configurations and F=[0,finit]∣G∣⊂R≥0∣G∣F = [0,f_{\mathrm{init}}]^{|G|} \subset\mathbb{R}_{\ge0}^{|G|} is the compact set of ECM concentrations. The global system, with state Sk=(σk,fk)S^k = (\sigma^k,f^k), satisfies the Markov property. Since FF is uncountable, {Sk}k∈N\{S^k\}_{k \in\mathbb{N}} is a discrete-time general state-space Markov chain in the sense of [32 Chapter 3]. We now characterize the absorbing states and establish convergence properties of the interconnected system. Specifically, the following theorem extends results from [1] on standard CPMs to the ECPM framework with ECM dynamics.

Theorem IV.3 (Global system convergence). The general state-space Markov chain {Sk}k∈N\{S^k\}_{k \in\mathbb{N}} represented by the ECPM dynamics in (15)-(16) converges almost surely to the equilibrium (absorbing) set A\mathcal{A}, given by

A=Aℓ0∪(⋃ℓ∈L∖{ℓ0}Aℓ)\mathcal{A} = \mathcal{A}_{\ell_0} \cup\left(\bigcup_{\ell\in L \setminus\{\ell_0\}} \mathcal{A}_{\ell}\right)

in which

Aℓ0:={S:σp=ℓ0, ∀p∈G},Aℓ:={S:σp=ℓ, fp=0, ∀p∈G}.\mathcal{A}_{\ell_0} := \{S : \sigma_p = \ell_0,\ \forall p \in G\},\qquad\mathcal{A}_{\ell} := \{S : \sigma_p = \ell,\ f_p = 0,\ \forall p \in G\}.

Specifically, the σ\sigma-dynamics converge with probability 11 in finite time to the uniform configuration set Aσ\mathcal{A}_{\sigma}, independently of ff and the initial condition S0S^0, where

Aσ:={σ∈L∣G∣:σp=σq, ∀(p,q)∈E}.\mathcal{A}_{\sigma} := \{\sigma\in L^{|G|} : \sigma_p = \sigma_q,\ \forall(p,q) \in\mathcal{E}\}.

Proof. See Appendix D.

Invasion Time in One-Dimensional ECPM

In this section, we analyze the ECPM in a one-dimensional (1D) lattice, focusing on the invasion time, i.e. the time steps required for the tumor front to reach a specific lattice site given an initial configuration. As highlighted in [33], the 1D-ECPM can be viewed as a reduced representation of tumor invasion, capturing migration dynamics that closely mirror the radial progression observed in three-dimensional settings. Moreover, we restrict admissible σ\sigma-updates to forward invasion, neglecting reverse copy attempts. This assumption is motivated as follows. Biologically, 1D migration along narrow ECM fibers exhibits stable polarity with predominant forward motion driven by matrix degradation at the leading edge [33]. Analytically, this restriction provides a conservative estimate: by preventing empty sites from overwriting tumor cells, we obtain a lower bound on the actual mean invasion time. Let GG be such that nx=1n_x = 1 and ny=nn_y = n, with nn being the number of lattice sites. Let the initial ECM density be uniform across the lattice, i.e.

fp0=finit,∀p∈G,(25)f_p^0 = f_{\mathrm{init}}, \qquad\forall p \in G, \tag*{(25)}

where finit∈R>0f_{\mathrm{init}} \in\mathbb{R}_{>0} denotes the initial ECM density at a lattice site. We denote by f0f^0 the initial ECM vector satisfying (23). It follows that fpk∈[0,finit]f_p^k \in[0,f_{\mathrm{init}}] for all k∈Nk \in\mathbb{N} and p∈Gp \in G (see Lemma IV.1). Now, we define the invasion time as follows.

Definition V.1 (Invasion time). Given a grid point j∈Gj \in G, we define the maximum tumor depth at iteration kk as φmax⁡(σk):=max⁡{yp∈Y:σpk≠ℓ0}\varphi_{\max}(\sigma^k) := \max\{y_p \in\mathcal{Y} : \sigma_p^k \ne\ell_0\}, where ypy_p denotes the vertical coordinate of p∈Gp \in G. The invasion time TjINVT_j^{\mathrm{INV}} is defined as the minimum number of ECPM update steps required for the tumor front to reach the lattice site at depth jj, i.e.

TjINV:=min⁡{k≥0:φmax⁡(σk)≥j}.T_j^{\mathrm{INV}} := \min\{k \ge0 : \varphi_{\max}(\sigma^k) \ge j\}.

We now introduce a compact formulation of the invasion time probability. Let Pj∣i(k∣S0)\mathcal{P}_{j \mid i}(k \mid S^0) indicate the probability that the tumor reaches lattice site jj in kk steps, starting from initial ECPM state S0S^0 with maximum tumor depth i≤ji \le j. Formally,

Pj∣i(k∣S0):=P(TjINV=k∣φmax⁡(σ0)=i, f0).(26)\mathcal{P}_{j \mid i}(k \mid S^0) := P\left(T_j^{\mathrm{INV}} = k \mid\varphi_{\max}(\sigma^0) = i,\ f^0\right). \tag*{(26)}

Building on Definition V.1, the mean invasion time is defined as the conditional expectation of TjINVT_j^{\mathrm{INV}} given the initial ECPM state. Since TjINVT_j^{\mathrm{INV}} is a discrete random variable taking values k∈Z>0k \in\mathbb{Z}_{>0} with probability mass function (24), we obtain

T‾j∣iINV(S0):=E[TjINV∣φmax⁡(σ0)=i, f0]=∑k=1∞k Pj∣i(k∣S0).(27)\overline{T}_{j \mid i}^{\mathrm{INV}}(S^0) := \mathbb{E}\left[T_j^{\mathrm{INV}} \mid\varphi_{\max}(\sigma^0) = i,\ f^0\right] = \sum_{k=1}^{\infty} k\,\mathcal{P}_{j \mid i}(k \mid S^0). \tag*{(27)}

provided that the series converges. In practice, the mean invasion time can be approximated by truncating the infinite series in (25) to a finite horizon K∈NK \in\mathbb{N}, yielding

T‾j∣iINV(S0)≈∑k=1Kk Pj∣i(k∣S0).(28)\overline{T}_{j \mid i}^{\mathrm{INV}}(S^0) \approx\sum_{k=1}^{K} k\,\mathcal{P}_{j \mid i}(k \mid S^0). \tag*{(28)}

We now proceed with the derivation of closed-form expressions for the invasion time probability Pj∣i(k∣S0)\mathcal{P}_{j\mid i}\left(k\mid S^{0}\right) for N≥1N\geq1 tumor cells. Let us consider a 1D-ECPM initialized in state S0S^{0}, as depicted in Figure 4. In detail, the initial configuration consists of NN contiguous tumor cells, labeled ℓ1,…,ℓN\ell_{1},\ldots,\ell_{N}, occupying the lattice GG. The tumor front is initially at position φmax⁡(σ0)=i\varphi_{\max}\left(\sigma^{0}\right)=i, and each cell ℓ\ell has an initial volume Vℓ0V_{\ell}^{0}. In this framework, the label and type sets are L={ℓ0,ℓ1,…,ℓN}L=\{\ell_{0},\ell_{1},\ldots,\ell_{N}\} and Γ={γECM,γc}\Gamma=\{\gamma_{ECM},\gamma_{c}\}, where ℓ0\ell_{0} corresponds to empty lattice sites covered by ECM, γECM\gamma_{ECM} and γc\gamma_{c} denote the ECM and tumor cell types respectively, with τℓ0=γECM\tau_{\ell_{0}}=\gamma_{ECM} and τℓ=γc\tau_{\ell}=\gamma_{c} for all ℓ≠ℓ0\ell\ne\ell_{0}. Moreover, since only tumor cells secrete degrading enzymes

Figure 4. Illustration of the initial ECPM state for a 1D grid with NN tumor cells surrounded by ECM. The indices ii, j>ij>i, and nn denote the initial front depth, target depth, and grid length, respectively.

(i.e. cγECMf=0c_{\gamma_{ECM}}^{f}=0), the dynamics in (5) simplify to

fpk+1=(1−cγcfz(p,γc,σk))fpk,∀p∈G,(29)f_{p}^{k+1}=\left(1-c_{\gamma_{c}}^{f}z\left(p,\gamma_{c},\sigma^{k}\right)\right)f_{p}^{k},\qquad\forall p\in G, \tag*{(29)}

with the proteolytic enzyme secreted at site pp (see general expression in (4)), given by

z(p,γc,σ)=cγcz∑d=1Dγcnd(p,γc,σ)e−αγcd.z\left(p,\gamma_{c},\sigma\right)=c_{\gamma_{c}}^{z}\sum_{d=1}^{D_{\gamma_{c}}}n_{d}\left(p,\gamma_{c},\sigma\right)e^{-\alpha_{\gamma_{c}}d}.

Before presenting the main result, we introduce some definitions needed to express the invasion time probability in a compact form. We denote by

b(k,j,i):=(k−1j−i−1),(30)b(k,j,i):=\binom{k-1}{j-i-1}, \tag*{(30)}

the number of ordered partitions of kk steps into j−ij-i forward transitions of the tumor front, namely transitions such that φmax⁡(σk)=i+s\varphi_{\max}\left(\sigma^{k}\right)=i+s for some s∈{1,…,j−i}s\in\{1,\ldots,j-i\}. Each partition is indexed by r∈{1,…,b(k,j,i)}r\in\{1,\ldots,b(k,j,i)\} and denoted by (k1r,…,kj−ir)\left(k_{1}^{r},\ldots,k_{j-i}^{r}\right), such that

k=∑s=1j−iksr,ksr≥1,(31)k=\sum_{s=1}^{j-i}k_{s}^{r},\qquad k_{s}^{r}\geq1, \tag*{(31)}

in which the index ss labels the successive front transitions. Additionally, within each partition rr we define the cumulative iteration counts ζsr:=∑g=1skgr\zeta_{s}^{r}:=\sum_{g=1}^{s}k_{g}^{r}, so that the sths^{\mathrm{th}} transition in the rthr^{\mathrm{th}} partition occurs between iterations ζs−1r\zeta_{s-1}^{r} and ζsr−1\zeta_{s}^{r}-1. The following theorem characterizes the invasion time for the setting introduced above.

Theorem V.2 (Invasion Time Probability for NN Tumor Cells in a 1D-ECPM). Consider a 1D lattice with an initial configuration consisting of N≥1N\geq1 contiguous tumor cells, where the tumor front is at starting position φmax⁡(σ0)=i\varphi_{\max}\left(\sigma^{0}\right)=i and let j>ij>i be a target front position. Then, the probability that the tumor front reaches site jj from ii in exactly k≥j−ik\geq j-i steps is given by

Pj∣i(k∣S0)=∑r=1b(k,j,i)∏s=ij−1Ps+1∣s(ks−i+1r∣Ss−i).(32)\mathcal{P}_{j\mid i}\left(k\mid S^{0}\right)=\sum_{r=1}^{b(k,j,i)}\prod_{s=i}^{j-1}\mathcal{P}_{s+1\mid s}\left(k_{s-i+1}^{r}\mid S^{s-i}\right). \tag*{(32)}

Here, each factor φsr\varphi_{s}^{r} corresponds to the probability of the (s−i+1)th(s-i+1)^{\mathrm{th}} transition within the rthr^{\mathrm{th}} partition to occur, i.e.

φsr=P(⋅∣Sζs−1r)2N−1∏h=ζs−1rζsr−1(1−P(⋅∣Sh)2N−1),(33)\varphi_{s}^{r}=\frac{P\left(\cdot\mid S^{\zeta_{s-1}^{r}}\right)}{2N-1}\prod_{h=\zeta_{s-1}^{r}}^{\zeta_{s}^{r}-1}\left(1-\frac{P\left(\cdot\mid S^{h}\right)}{2N-1}\right), \tag*{(33)}

where, at each intermediate step hh, P(⋅∣Sh)P\left(\cdot\mid S^{h}\right) is the transition probability in (19), with energy variation defined by

ΔE(Sh)=λ[1+2(VℓNh−VT)]+KECM(fph−fqh),(34)\Delta E\left(S^{h}\right)=\lambda\left[1+2\left(V_{\ell_{N}}^{h}-V_{T}\right)\right]+K_{ECM}\left(f_{p}^{h}-f_{q}^{h}\right), \tag*{(34)}

where pp and qq denote, respectively, the current front position φmax⁡(σh)\varphi_{\max}\left(\sigma^{h}\right) and its adjacent empty-site neighbor.

Proof. See Appendix E.

Numerical Simulations and Validation

In this section, we validate the analytical results from § 5 via numerical experiments. We begin by detailing the simulation setup and parameter choices. Next, we conduct a sensitivity analysis of the mean invasion time, comparing empirical results against closed-form solutions, across key ECPM parameters and various invasion distances.

Simulation Details

All simulations were implemented in Python programming language by modeling the coupled tumor-ECM dynamics described in § 3. Each run starts from an initial configuration S0\mathcal{S}^{0}, corresponding to the framework illustrated in Figure 4. To choose S0\mathcal{S}^{0}, we resorted to typical experimental setups in which tumor spheroids are seeded in a collagen matrix. Specifically, recalling that 1D scenarios mimic the cross-section of a 3D tumor spheroid, we considered an initial state in which a contiguous block of cells NN occupies 50% of the 1D-lattice, for instance N=3N=3 cells with initial volumes V0=10V_{0}=10 in a lattice of length n=60n=60.

We now specify the ECPM parameter values used in the simulations. The environmental noise parameter in (19), weighing the probability of energetically unfavorable events, was set to β=4\beta=4. We set the target cell volume to VT=2V0V_{T}=2V_{0} (see (12)) to allow for significant morphological expansion during invasion. This choice provides the cell with sufficient volumetric freedom to deform, elongate, and invade the surrounding medium without being overly constrained by the volume conservation term in the Hamiltonian. Adhesion energies were chosen as in (7), specifically Jc,c=3J_{c,c}=3 and JECM,c=6J_{\mathrm{ECM},c}=6 considering a stronger adhesion between tumor cells and ECM than between tumor cells themselves (typical of invasive phenotypes). However, since the invasion time probability in Theorem V.2 does not depend on adhesion energies, these values do not affect the analytical results. The enzyme decay rate was set to aγc=2a_{\gamma_{c}}=2, while the product between the secretion and ECM degradation rates in (5) was set to cγcz⋅cγcf=0.01c_{\gamma_{c}}^{z}\cdot c_{\gamma_{c}}^{f}=0.01, and the maximal influence radius to Dγc=4D_{\gamma_{c}}=4 (see (4)). The adopted ECPM parameter set is consistent with standard ECPM configurations reported in the literature [3, 10, 11], suitably adapted to the one-dimensional framework for validation purposes.

To estimate the mean invasion time T‾j∣iINV\overline{T}_{j\mid i}^{INV} we performed 10 000 Monte Carlo simulations of the ECPM dynamics in (17)–(18), adopting the stopping criterion φmax⁡(σk)=j\varphi_{\max}(\sigma^{k})=j. Here, jj denotes the target lattice site reached by the tumor front, while ii represents its initial position (i.e., φmax⁡(σ0)=i\varphi_{\max}(\sigma^{0})=i). Each simulation terminates when the front reaches site j>ij>i, and the corresponding invasion time Tj∣iINVT_{j\mid i}^{INV} is recorded. Moreover, to control computational cost, we set a maximum number of iteration steps K∈NK\in\mathbb{N}. This horizon is chosen to capture most of the probability mass of the invasion-time distribution, ensuring accurate estimates of the mean invasion time. Details on the stopping criterion used to select KK are provided in the next subsection.

Sensitivity of Invasion Time to λ\lambda and KECMK_{\mathrm{ECM}}

We now investigate how the mean invasion time T‾j∣iINV(S0)\overline{T}_{j\mid i}^{INV}(\mathcal{S}^{0}) scales with the invasion distance j−ij-i, and how this scaling is shaped by the two key parameters entering the energy term (34), namely the volume-constraint weight λ\lambda and the haptotactic coupling gain KECMK_{\mathrm{ECM}}. To validate the closed-form expression established in Theorem V.2 against empirical methods, for every parameter configuration we compare three mean invasion time estimators:

  • Empirical (full model): sample mean over Monte Carlo runs of the ECPM dynamics defined in § 3, in which ECM-over-tumor interactions are bidirectional, and the tumor front can advance or retreat at each step;

  • Empirical (simplified model): sample mean over Monte Carlo runs of the forward-only restricted dynamics introduced in the analysis of § 5;

  • Closed-form (simplified model): the truncated analytical mean of (28), evaluated via Theorem V.2.

For each parameter configuration, the truncation horizon KK used to evaluate (28) is selected via a practical heuristic. We therefore choose KK as the smallest integer satisfying cumulative-probability and point-mass thresholds for both the empirical and closed-form distributions, i.e.

∑k=1KPj∣i(k∣S0)≥0.98,Pj∣i(K∣S0)<10−5,(35)\sum_{k=1}^{K}\mathcal{P}_{j\mid i}(k\mid\mathcal{S}^{0})\geq0.98,\qquad\mathcal{P}_{j\mid i}(K\mid\mathcal{S}^{0})<10^{-5}, \tag*{(35)}

in which the second condition is checked only after a sufficiently large KK ensuring that the majority of the probability mass is captured. This guarantees a comparable level of distributional coverage across all the parameter combinations considered below, even though the resulting KK value itself varies substantially with λ\lambda, KECMK_{\mathrm{ECM}}, and the invasion distance, as shown next. We analyze the relative influence of λ\lambda and KECMK_{\mathrm{ECM}} on the mean invasion time. Figure 5 reports T‾j∣iINV\overline{T}_{j\mid i}^{INV} as a function of the invasion distance for several values of λ\lambda; the shaded band around each curve shows the family of functions obtained by varying KECM∈[1,20]K_{\mathrm{ECM}}\in[1,20]. Figure 6 provides the complementary view, grouping curves by KECMK_{\mathrm{ECM}} and showing shaded bands for λ∈[1,5]\lambda\in[1,5]. The closed-form estimator is independent of both λ\lambda and KECMK_{\mathrm{ECM}}, and it coincides with the empirical estimator obtained from the simplified model, thereby validating our analytical derivation. Notably, estimation quality decreases with increasing invasion distance, which is consistent with a higher incidence of tumor retreat events, that explains the growing discrepancy between estimators for large invasion distances. To assess the accuracy of the closed-form expression from another viewpoint, Figure 7 reports the three estimators introduced above, for three representative values λ∈{0.5,1,5}\lambda\in\{0.5, 1, 5\}, together with the corresponding truncation horizon KK satisfying conditions (33).

Mean invasion time estimations versus invasion distance for several values of $\lambda$, with shaded variance regions

Figure 5. Mean invasion time estimations versus invasion distance j−ij-i for several values of λ\lambda. Shaded areas denote the variance induced by varying KECMK_{ECM} around each λ\lambda curve.

Mean invasion time estimations versus invasion distance for several values of $K_{ECM}$, with shaded variance regions

Figure 6. Mean invasion time estimations versus invasion distance j−ij-i for several values of KECMK_{ECM}. Shaded areas denote the variance induced by varying λ\lambda around each KECMK_{ECM} curve.

Mean invasion time versus invasion distance for different $\lambda$ values

Figure 7. Mean invasion time versus invasion distance j−ij-i for different λ\lambda values, comparing the empirical (full model), empirical (simplified model), and closed-form estimators. The right axis (gray dashed line) reports the truncation horizon KK satisfying (35) for each invasion distance.

Conclusions

In this paper, we introduced a system-theoretic framework for analyzing an ECPM for tumor invasion by reformulating it as an interconnected system that couples a stochastic process describing locally interacting tumor cells with a jump-linear system modeling the ECM environment. Within this formulation, we characterized the asymptotic behavior of both subsystems, derived transition probabilities and uniform bounds, and exploited the sparse structure of the energy variation terms. These results yielded global convergence guarantees, showing that the coupled dynamics reach equilibrium states independently of the ECM field. In one-dimensional settings, we derived a closed-form for invasion-time probabilities that depend on local ECM gradients, number of cells and target cell volume. Numerical simulations validated the theory across multiple scenarios and supported the sensitivity analysis of the mean invasion time with respect to key ECPM parameters and invasion distances, providing an analytical foundation for tumor invasion dynamics in ECM substrates.

Appendix

Proof of Lemma III.1

The adhesion-energy change ΔHJ=HJ(σ′)−HJ(σ)\Delta H_J = H_J(\sigma')-H_J(\sigma) is computed considering all pairwise interactions between neighboring sites (see (11)). However, since only the interaction terms involving site qq and thus p∈Nqp \in\mathcal{N}_q change, all other terms cancel out, leading to equation (18). For the volume-energy change ΔHV=HV(σ′)−HV(σ)\Delta H_V = H_V(\sigma')-H_V(\sigma), after the copy attempt from pp to qq, the affected cell volumes evolve according to Vℓ+=Vℓ+ρ(σp,ℓ)V_\ell^{+}=V_\ell+\rho(\sigma_p,\ell), ∀ℓ∈{σp,σq}\forall\ell\in\{\sigma_p,\sigma_q\}, where ρ(ℓ1,ℓ2)=2δ(ℓ1,ℓ2)−1\rho(\ell_1,\ell_2)=2\delta(\ell_1,\ell_2)-1 accounts for the increase or decrease in volume. All the other cell volumes remain unchanged, thus canceling out in the energy difference. Therefore, the Hamiltonian volume energies at σ\sigma and σ′\sigma' can be simplified as

HV(σ)=λ∑ℓ∈{σp,σq}(Vℓ−VT)2H_V(\sigma)=\lambda\sum_{\ell\in\{\sigma_p,\sigma_q\}}(V_\ell-V_T)^2
HV(σ′)=λ∑ℓ∈{σp,σq}(Vℓ+ρ(σp,ℓ)−VT)2H_V(\sigma')=\lambda\sum_{\ell\in\{\sigma_p,\sigma_q\}}(V_\ell+\rho(\sigma_p,\ell)-V_T)^2

Then, subtracting these two expressions, developing the squares, and simplifying leads to the following volume-energy change

ΔHV(σ)=λ∑ℓ∈{σp,σq}[ρ(σp,ℓ)2+2ρ(σp,ℓ)(Vℓ−VT)]=λ∑ℓ∈{σp,σq}[1+2ρ(σp,ℓ)(Vℓ−VT)].\Delta H_V(\sigma)=\lambda\sum_{\ell\in\{\sigma_p,\sigma_q\}}\left[\rho(\sigma_p,\ell)^2+2\rho(\sigma_p,\ell)(V_\ell-V_T)\right]=\lambda\sum_{\ell\in\{\sigma_p,\sigma_q\}}\left[1+2\rho(\sigma_p,\ell)(V_\ell-V_T)\right].

which corresponds to equation (19) where the empty label ℓ0\ell_0 is excluded through the term (1−δ(ℓ0,ℓ))(1-\delta(\ell_0,\ell)). The ECM-dependent term KECM(fp−fq)K_{\mathrm{ECM}}(f_p-f_q) arises directly from definition 11 in Algorithm 1, thus completing the decomposition of ΔE(σ,f)\Delta E(\sigma,f) stated in the Lemma.

Proof of Lemma IV.1

Considering the dynamics of fpkf_p^k, we have for all p∈Gp\in G that

0≤fpk+1=Ap(σk)fpk≤fpk≤fp0,0\le f_p^{k+1}=A_p(\sigma^k)f_p^k\le f_p^k\le f_p^0,

Hence, by the monotone convergence theorem the sequence admits a limit in the interval [0,fp0][0,f_p^0], ∀p∈G\forall p\in G. When σk=σˉ\sigma^k=\bar{\sigma} for all k∈Nk\in\mathbb{N}, the diagonal structure of A(σˉ)A(\bar{\sigma}) yields fpk=[Ap(σˉ)]kfp0f_p^k=[A_p(\bar{\sigma})]^k f_p^0. The stated limit follows since Ap(σˉ)∈[0,1]A_p(\bar{\sigma})\in[0,1] determines whether the sequence decays to zero or remains constant.

Proof of Lemma IV.2

Consider a trial cell mass configuration σ′\sigma' such that σq′=σp\sigma'_q=\sigma_p, with p∈Gp\in G and q∈Npq\in\mathcal{N}_p. From the proof of Lemma IV.1, it follows that the ECM variation satisfies (fpk−fqk)∈[−finit,finit](f_p^k-f_q^k)\in[-f_{\mathrm{init}},f_{\mathrm{init}}], ∀k≥1\forall k\ge1, where finitf_{\mathrm{init}} is the initial uniform ECM concentration. By definition of ΔHmax⁡\Delta H_{\max} and ΔHmin⁡\Delta H_{\min}, the total energy variation can be bounded as follows

ΔE∈[ΔHmin⁡−KECMfinit,ΔHmax⁡+KECMfinit].(36)\Delta E\in[\Delta H_{\min}-K_{\mathrm{ECM}}f_{\mathrm{init}},\Delta H_{\max}+K_{\mathrm{ECM}}f_{\mathrm{init}}]. \tag*{(36)}

To establish the denominator, we calculate the total number of ordered neighboring pairs in E\mathcal{E}. Each interior point (x,y)(x,y), namely x∈Xx \in\mathcal{X} and y∈{2,…,ny−1}y \in\{2,\ldots,n_y-1\}, has 8 nearest neighbors. Instead, the top/bottom boundary points at y=1y=1 and y=nyy=n_y each have nxn_x points with 5 nearest neighbors. It follows that ∣E∣=nx(ny−2)⋅8+2nx⋅5=nx(8ny−6)|\mathcal{E}|=n_x(n_y-2)\cdot8+2n_x\cdot5=n_x(8n_y-6) which reduces to ∣E∣=nx(ny−2)⋅5+2nx⋅3=nx(5ny−4)|\mathcal{E}|=n_x(n_y-2)\cdot5+2n_x\cdot3=n_x(5n_y-4) and ∣E∣=nx(ny−2)⋅2+2nx⋅1=nx(2ny−2)|\mathcal{E}|=n_x(n_y-2)\cdot2+2n_x\cdot1=n_x(2n_y-2) for nx=2n_x=2 and nx=1n_x=1, respectively. These scenarios can be included by defining m:=min⁡(nx,3)m:=\min(n_x,3) to then obtain ∣E∣=nx[(ny−2)(3m−1)+2(2m−1)]|\mathcal{E}|=n_x[(n_y-2)(3m-1)+2(2m-1)]. Note that, in the Lemma statement, we use the upper bound ∣E∣≤nx(8ny−6)|\mathcal{E}|\leq n_x(8n_y-6) for simplicity. From expressions (20) and (21), a σ\sigma-dependent lower bound on the transition probability would involve the denominator ∣E∣−NU(σ)|\mathcal{E}|-N_U(\sigma), since only edges with differing labels are feasible. To obtain an absolute lower bound we consider the worst case NU(σ)=0N_U(\sigma)=0, i.e., all neighboring pairs carry distinct labels, yielding the largest possible denominator ∣E∣=nx(8ny−6)|\mathcal{E}|=n_x(8n_y-6). Substituting the worst-case energy increase and denominator in (21) directly yields the stated lower bound for any σ′∈Σσ+(σ,f)\sigma' \in\Sigma^+_\sigma(\sigma,f). The same bound applies to any σ′∈Σσ−(σ,f)\sigma' \in\Sigma^-_\sigma(\sigma,f) since in that case the transition probability (20) is higher. For the upper bound, observe that NU(σ)≤∣E∣−2N_U(\sigma)\leq|\mathcal{E}|-2 always holds, since at least two edges with different labels exist unless the configuration is absorbing. Thus, the denominator is at least 2, yielding the uniform upper bound 1/21/2 for the transition probability. Precisely, the upper bound is 1/φ1/\varphi with φ≥2\varphi\geq2 depending on the specific boundary conditions and metric defined on GG.

Proof of Theorem IV.3

We begin by identifying the equilibria of the ECPM dynamics in (15)-(16), which correspond to the absorbing states of the Markov chain {Sk}k∈N\{S^k\}_{k\in\mathbb{N}}.

Case 1 (σp=ℓ0\sigma_p=\ell_0, ∀p∈G\forall p\in G): When all lattice sites carry the empty label ℓ0\ell_0, no tumor cell occupies any site. By definition of the enzyme model (4), cells of type γECM\gamma_{ECM} produce no degrading enzyme (i.e., cγECMf=0c^{f}_{\gamma_{ECM}}=0), so Ap(σ)=1A_p(\sigma)=1 for all p∈Gp\in G. Consequently, the ECM field remains constant at its current value fp∈[0,fp0]f_p\in[0,f_p^0], yielding the absorbing subset Aℓ0\mathcal{A}_{\ell_0}.

Case 2 (σp=ℓ≠ℓ0\sigma_p=\ell\neq\ell_0, ∀p∈G\forall p\in G): When every lattice site is occupied by a tumor cell of label ℓ\ell, the degradation factor satisfies Ap(σ)<1A_p(\sigma)<1 for all p∈Gp\in G, since each site lies within the influence radius of at least one tumor cell, implying αp(σ)>0\alpha_p(\sigma)>0 in (14). Therefore, by the fixed-configuration case in Lemma IV.1, the ECM concentration at every lattice point decays monotonically to zero. Once f=0∣G∣f=0_{|G|}, the ECM field is frozen and the configuration (σ,0∣G∣)(\sigma,0_{|G|}) is absorbing, yielding the subset Aℓ\mathcal{A}_{\ell}.

We now show that the system trajectories converge almost surely to the absorbing set A\mathcal{A}. By Lemma IV.1, fkf^k converges regardless of σ\sigma, allowing us to treat the σ\sigma-dynamics as a time-varying process whose transition probabilities evolve through fkf^k. Therefore, the cumulative transition matrix of the σ\sigma-dynamics up to step kk admits the canonical block form

(∏s=0kQsR~k0I),R~k:=∑h=0k(∏s=0h−1Qs)Rh,\begin{pmatrix} \prod_{s=0}^{k}Q_s & \widetilde{R}_k\\ 0 & I \end{pmatrix}, \qquad \widetilde{R}_k:=\sum_{h=0}^{k}\left(\prod_{s=0}^{h-1}Q_s\right)R_h,

where QsQ_s is the sub-stochastic transition matrix among transient states at step ss, RhR_h is the transition matrix from transient to absorbing states at step hh, and II is the identity on absorbing states. The sum of the elements of the ii-th row of the matrix

B(t,m):=∏k=tt+m−1QkB(t,m):=\prod_{k=t}^{t+m-1}Q_k

represents the probability of not being absorbed for mm steps starting from transient state ii at time tt. Since, by Lemma IV.2 each 1-step non-zero transition probability is lower bounded by the positive constant ϱ\varrho, and that, for each transient state ii, there exists an admissible path of length mim_i to an absorbing state, if we set M:=max⁡i{mi}M:=\max_i\{m_i\}, we have

∥B(t,M)∥∞=max⁡i∑j[B(t,M)]ij≤1−ϱM.(37)\lVert B(t,M)\rVert_\infty=\max_i\sum_j[B(t,M)]_{ij}\leq1-\varrho^M. \tag*{(37)}

Then, since the iteration step can be decomposed as k=M⌊k/M⌋+k mod Mk=M\lfloor k/M\rfloor+k\bmod M, we can write

∥∏s=0kQs∥∞≤(a)(∏s=0⌊k/M⌋−1∏s=0M−1∥QMs+t∥∞)∏s=⌊k/M⌋Mk∥Qs∥∞=(∏s=0⌊k/M⌋−1∥B(Ms,M)∥∞)∏s=⌊k/M⌋Mk∥Qs∥∞\left\lVert\prod_{s=0}^{k}Q_s\right\rVert_\infty \overset{(a)}{\leq} \left(\prod_{s=0}^{\lfloor k/M\rfloor-1}\prod_{s=0}^{M-1}\lVert Q_{Ms+t}\rVert_\infty\right) \prod_{s=\lfloor k/M\rfloor M}^{k}\lVert Q_s\rVert_\infty = \left(\prod_{s=0}^{\lfloor k/M\rfloor-1}\lVert B(Ms,M)\rVert_\infty\right) \prod_{s=\lfloor k/M\rfloor M}^{k}\lVert Q_s\rVert_\infty
≤(b)(1−ϱM)⌊k/M⌋∏s=⌊k/M⌋Mk∥Qs∥∞≤(c)(1−ϱM)⌊k/M⌋(n2)M,\overset{(b)}{\leq} (1-\varrho^M)^{\lfloor k/M\rfloor} \prod_{s=\lfloor k/M\rfloor M}^{k}\lVert Q_s\rVert_\infty \overset{(c)}{\leq} (1-\varrho^M)^{\lfloor k/M\rfloor}\left(\frac{n}{2}\right)^M,

where in (a) we used the Schwarz inequality on the infinity operator norm, in (b) we applied the bound in (34), and in (c), where we denote by n:=∣Lnx×ny∣−∣A∣n:=|L^{n_x\times n_y}|-|\mathcal{A}| the total number of transient configurations, we leveraged the upper bound on the transition probabilities (cf. Lemma IV.2) and the fact that k mod M≤Mk\bmod M\leq M which implies that the second term in (c) consists of at most MM factors. It follows that the probability of not being absorbed in kk steps, starting from an arbitrary transient state, goes to zero as k→∞k\to\infty. Then, the probability of being absorbed in finite time is 1, regardless of S0S^0.

Proof of Theorem V.2

Consider the 1D-ECPM with NN contiguous cells and tumor front at position ii (Figure 4). Under the forward-invasion restriction, feasible transitions divide into:

  • Internal transitions: adjacent cells invade each another through one of the 2(N−1)2(N-1) internal fronts without changing the front position. The resulting configurations are thus given by Σaint:={σ′:φmax⁡(σ′)=i}\Sigma_a^{\mathrm{int}} := \{\sigma' : \varphi_{\max}(\sigma') = i\}

  • Front-advance transition: front cell moves to site i+1i+1, leading to the (unique) configuration, represented by the singleton set Σafront:={σ′:φmax⁡(σ′)=i+1}\Sigma_a^{\mathrm{front}} := \{\sigma' : \varphi_{\max}(\sigma') = i+1\},

With 2N−12N-1 total feasible configurations selected with equal probability, the probability of selecting the front-advance configuration σ′∈Σafront\sigma' \in\Sigma_a^{\mathrm{front}} is 12N−1\frac{1}{2N-1}. Then, we compute Pi+1∣i(k∣S0)\mathcal{P}_{i+1\mid i}(k\mid S^0) and the associated energy change ΔE(Sh)\Delta E(S^h) at each step h∈{0,…,k−1}h \in\{0,\ldots,k-1\} of the next-site invasion process, leveraging Lemma III.1. By the sparse decomposition in (20), the adhesion energy change for site i+1i+1 yields

ΔHJ=∑w∈Ni+1[J~(σi,σw)−J~(σi+1,σw)]=J~(ℓN,ℓN)−J~(ℓ0,ℓN)+J~(ℓN,ℓ0)−J~(ℓ0,ℓ0)=0,\Delta H_J = \sum_{w\in\mathcal{N}_{i+1}}\left[\widetilde{J}(\sigma_i,\sigma_w)-\widetilde{J}(\sigma_{i+1},\sigma_w)\right] = \widetilde{J}(\ell_N,\ell_N)-\widetilde{J}(\ell_0,\ell_N)+\widetilde{J}(\ell_N,\ell_0)-\widetilde{J}(\ell_0,\ell_0)=0,

that cancels out based on the standard choice of the adhesion energy function in (7). This means that the front advance in 1D-ECPM is independent of the adhesion energy. Instead, for the volume energy, only the front cell evolves, yielding the following change ΔHV=λ[1+2(VℓN−VT)]\Delta H_V=\lambda[1+2(V_{\ell_N}-V_T)]. At the first step, uniform ECM initialization in (25) makes the degradation term vanish, leading to ΔE(S0)=λ[1+2(VℓN0−VT)]\Delta E(S^0)=\lambda[1+2(V_{\ell_N}^0-V_T)]. For h∈{1,…,k−2}h\in\{1,\ldots,k-2\} considering a fixed tumor configuration σ0\sigma^0 and evolving ECM, the associated energy is

ΔE(Sh)=λ[1+2(VℓNh−VT)]+KECM(fph−fqh),(38)\Delta E(S^h)=\lambda[1+2(V_{\ell_N}^h-V_T)]+K_{\mathrm{ECM}}(f_p^h-f_q^h), \tag*{(38)}

where, by Lemma IV.1, the ECM term KECM(fph−fqh)K_{\mathrm{ECM}}(f_p^h-f_q^h) is negative for any hh, progressively reducing the energy. Then, the front-advance transition probability to σh+1∈Σafront\sigma^{h+1}\in\Sigma_a^{\mathrm{front}} at next step is obtained by combining the front selection probability with the Boltzmann transition probability in (19), giving

P(σh+1∣Sh)=12N−1{e−ΔEh/βif ΔE(Sh)>0,1otherwise.(39)P(\sigma^{h+1}\mid S^h)=\frac{1}{2N-1} \begin{cases} e^{-\Delta E^h/\beta} & \text{if }\Delta E(S^h)>0,\\ 1 & \text{otherwise} \end{cases} . \tag*{(39)}

Note that, the front advance is energetically unfavorable when ΔE(Sh)>0\Delta E(S^h)>0, i.e. under ECPM parameter condition

VT−VℓNh≤12(KECM(fph−fqh)λ+1).(40)V_T-V_{\ell_N}^h\leq\frac{1}{2}\left(\frac{K_{\mathrm{ECM}}(f_p^h-f_q^h)}{\lambda}+1\right). \tag*{(40)}

Now, for exactly one site to advance in kk steps, the front must remain stationary for the first k−1k-1 steps considering either internal transitions σh∈Σaint\sigma^h\in\Sigma_a^{\mathrm{int}} or rejected advances σh∈Σafront\sigma^h\in\Sigma_a^{\mathrm{front}}, to then finally advance at step kk. This yields

Pi+1∣i(k∣S0)=P(σk∣Sk−1)∏h=0k−2ψh,(41)\mathcal{P}_{i+1\mid i}(k\mid S^0)=P(\sigma^k\mid S^{k-1})\prod_{h=0}^{k-2}\psi_h, \tag*{(41)}

where the front-advancing transition probability P(σk∣Sk−1)P(\sigma^k\mid S^{k-1}) is given by (39), and the terms ψh\psi_h represent the probability of not advancing at step hh, defined as

ψh:=2(N−1)2N−1+1−e−ΔE(Sh)/β2N−1.\psi_h:=\frac{2(N-1)}{2N-1}+\frac{1-e^{-\Delta E(S^h)/\beta}}{2N-1}.

Here, rearranging the terms and simplifying yields (33). We now extend the single-site result to multi-site progression from ii to jj. With j−ij-i successive advances needed, we decompose kk total steps via ordered partitions (k0r,…,kj−i−1r)(k_0^r,\ldots,k_{j-i-1}^r) satisfying (31), where each term specifies the timing of the ss-th transition within the rr-th partition. For each partition, the invasion probability factors as a product of single-site terms from equation (41). Summing over all partitions we obtain the following invasion probability

Pj∣i(k∣S0)=∑r=1b(k,j,i)∏s=ij−1δsr,\mathcal{P}_{j\mid i}(k\mid S^0)=\sum_{r=1}^{b(k,j,i)}\prod_{s=i}^{j-1}\delta_s^r,

where b(k,j,i)b(k,j,i) is defined as in (30), and δsr\delta_s^r is computed as in (41) considering the step indexes referred to the ss-th transition within the rr-th partition. This is the compact form in (32).

References

Biographies

Andrea Tramaloni

Andrea Tramaloni received the M.Sc. degree (2022) in Automation Engineering, and the Ph.D. in Biomedical, Electrical, and Systems Engineering (2026), from the University of Bologna, Italy. He was a visiting scholar at the University of Bristol, UK, in 2025. He is currently a Post-doc at CASY, Department of Electrical, Electronic, and Information Engineering G. Marconi of Alma Mater Studiorum Università di Bologna. His research interests include system biology, multi-agent systems, and control theory.

Simone Baroncini

Simone Baroncini received the B.Sc. degree (2019), the M.Sc. degree “summa cum laude” (2022) in Automation Engineering, and the Ph.D. in Biomedical, Electrical, and Systems Engineering (2026), from the University of Bologna, Italy. He was a Visiting Scholar at Queen’s University, Kingston, Ontario, Canada, in 2025. He is currently a Post-doc at CASY, Department of Electrical, Electronic, and Information Engineering G. Marconi of Alma Mater Studiorum Università di Bologna. His research interests are in the intersections of reinforcement learning, control theory, and geometry.

Sofia Avnet

Sofia Avnet received a Degree in Medical Biotechnologist, in 2000, and the Ph.D. degree, in 2005, a second Degree in cellular and molecular biology, in 2010, and the Specialization degree in pharmacology and toxicology, in 2023. During the first master’s degree, she began research on bone tumors with the Kimmel Cancer Institute and the IRCCS Istituto Ortopedico Rizzoli. For the Ph.D. degree, she continued studying biological mechanisms of bone cancers with the Institute for Cancer Research, Candiolo. She also conducted research on bone pathophysiology as a Visiting Researcher with McGill University, in 2006, and a Research Fellow with the University of Turku, in 2007. From 2008 to 2019, she was a Biotechnologist with the Rizzoli Orthopedic Institute, focusing her research on tumor invasiveness, chemoresistance, microenvironment, and metabolism in bone cancers. Since 2019, she has been an Assistant Professor with the University of Bologna, where she develops 3D bone tissue and bone cancer models using spheroids, microfluidics, and bioprinting. Over her career, she has received several research grants as a Principal Investigator from the Italian Association of Cancer Research and the Ministry of Health. Dr. Avnet has been the President of the International Society of Cancer Metabolism (2015-2016).

Nicola Baldini

Nicola Baldini received the Degree in medicine and surgery, the Certification degree in orthopaedics, and the Certification degree in oncology from the University of Bologna, Bologna, Italy, in 1981, 1987, and 1990, respectively. From 1988 to 1989, he was a Research Fellow with Harvard University. From 2013 to 2018, he was a Visiting Professor with Kyoto Prefectural University of Medicine. He is currently a Professor with the Department of Biomedical and Neuromotor Sciences, Alma Mater Studiorum, University of Bologna, and the Director of the Department of Research, Innovation and Technology, IRCCS Istituto Ortopedico Rizzoli, Bologna. His research interests include the microenvironment of bone tissue under physiological and pathological conditions, bone cancers, and regenerative medicine. Prof. Baldini has served as the President for the Connective Tissue Oncology Society, from 1999 to 2000, the European Orthopaedic Research Society, from 2008 to 2012, and the International Society for Cancer Metabolism, from 2019 to 2020.

Giuseppe Notarstefano

Giuseppe Notarstefano is a Professor in the Department of Electrical, Electronic, and Information Engineering G. Marconi at Alma Mater Studiorum Università di Bologna. He was Associate Professor (June ’16 - June ’18) and previously Assistant Professor, Ricercatore, (February ’07 - June ’16) at the Università del Salento, Lecce, Italy. He received the Laurea degree “summa cum laude” in Electronics Engineering from the Università di Pisa in 2003 and the Ph.D. degree in Automation and Operation Research from the Università di Padova in 2007. He has been visiting scholar at the University of Stuttgart, University of California Santa Barbara and University of Colorado Boulder. His research interests include distributed optimization, cooperative control in complex networks, applied nonlinear optimal control, trajectory optimization and maneuvering of aerial and car vehicles. He serves as an Associate Editor for IEEE Transactions on Automatic control, IEEE Transactions on Control Systems Technology and IEEE Control Systems Letters. He is also part of the Conference Editorial Board of IEEE Control Systems Society and EUCA. He is recipient of an ERC Starting Grant.

References

  1. [1]V. Voss-Böhme, “Multi-scale modeling in morphogenesis: a critical analysis of the cellular potts model,” PLoS One, 2012.
  2. [2]M. J. Oudin, O. Jonas, T. Kosciuk, L. C. Broye, B. C. Gujdo, J. Wyckoff, D. Riquelme, J. M. Lamar, S. B. Asokan, C. Whittaker et al., “Tumor cell–driven extracellular matrix remodeling drives haptotaxis during metastatic progression,” Cancer discovery, vol. 6, no. 5, pp. 516–531, 2016.DOI
  3. [3]F. Graner and J. A. Glazier, “Simulation of biological cell sorting using a two-dimensional extended potts model,” Physical review letters, vol. 69, no. 13, p. 2013, 1992.
  4. [4]V. Beffara and H. Duminil-Copin, “The self-dual point of the two-dimensional random-cluster model is critical for q ≥ 1,” Probability Theory and Related Fields, vol. 153, no. 3, pp. 511–542, 2012.
  5. [5]H. Duminil-Copin, V. Sidoravicius, and V. Tassion, “Continuity of the phase transition for planar random-cluster and potts models with 1 ≤ q ≤ 4,” Communications in Mathematical Physics, vol. 349, pp. 47–107, 2017.
  6. [6]U. A. Rozikov, Gibbs measures in biology and physics: The Potts model. World Scientific, 2022.DOI
  7. [7]M. Alber, N. Chen, P. M. Lushnikov, and S. A. Newman, “Continuous macroscopic limit of a discrete stochastic model for interaction of living cells,” Physical review letters, vol. 99, no. 16, p. 168102, 2007.DOI
  8. [8]D. Dan, C. Mueller, K. Chen, and J. A. Glazier, “Solving the advection-diffusion equations in biological contexts using the cellular potts model,” Physical Review E—Statistical, Nonlinear, and Soft Matter Physics, vol. 72, no. 4, p. 041909, 2005.pubmed.ncbi.nlm.nih.gov/16383422
  9. [9]N. J. Savill and P. Hogeweg, “Modelling morphogenesis: from single cells to crawling slugs,” Journal of theoretical biology, vol. 184, no. 3, pp. 229–235, 1997.DOI
  10. [10]J. Turner and J. A. Sherratt, “Intercellular adhesion and cancer invasion: a discrete simulation using the extended potts model,” Journal of theoretical biology, vol. 216, no. 1, pp. 85–100, 2002.pubmed.ncbi.nlm.nih.gov/12076130
  11. [11]S. Turner, J. A. Sherratt, and D. Cameron, “Tamoxifen treatment failure in cancer and the nonlinear dynamics of tgfβ,” Journal of theoretical biology, vol. 229, no. 1, pp. 101–111, 2004.pubmed.ncbi.nlm.nih.gov/15178188
  12. [12]A. Szabó, K. Varga, T. Garay, B. Hegedűs, and A. Czirók, “Invasion from a cell aggregate—the roles of active cell motion and mechanical equilibrium,” Physical biology, vol. 9, no. 1, p. 016010, 2012.DOI
  13. [13]J. T. Daub and R. M. Merks, “A cell-based model of extracellular-matrix-guided endothelial cell migration during angiogenesis,” Bulletin of mathematical biology, vol. 75, pp. 1377–1399, 2013.DOI
  14. [14]M. Scianna, “An extended cellular potts model analyzing a wound healing assay,” Computers in Biology and Medicine, vol. 62, pp. 33–54, 2015.DOI
  15. [15]T. Hirashima, E. G. Rens, and R. M. Merks, “Cellular potts modeling of complex multicellular behaviors in tissue morphogenesis,” Development, growth & differentiation, vol. 59, no. 5, pp. 329–339, 2017.DOI
  16. [16]R. M. Crossley, S. Johnson, E. Tsingos, Z. Bell, M. Berardi, M. Botticelli, Q. J. Braat, J. Metzcar, M. Ruscone, Y. Yin et al., “Modeling the extracellular matrix in cell migration and morphogenesis: a guide for the curious biologist,” Frontiers in cell and developmental biology, vol. 12, p. 1354132, 2024.DOI
  17. [17]J. P. Jansch-Porto, B. Hu, and G. E. Dullerud, “Policy optimization for Markovian jump linear quadratic control: Gradient-based method and global convergence,” IEEE Transactions on Automatic Control, vol. 68, no. 4, pp. 2475–2482, 2023.
  18. [18]M. Ogura and V. M. Preciado, “Optimal design of switched networks of positive linear systems via geometric programming,” IEEE Transactions on Control of Network Systems, vol. 4, no. 2, pp. 213–222, 2017.
  19. [19]P. E. Paré, C. L. Beck, and A. Nedíc, “Epidemic processes over time-varying networks,” IEEE Transactions on Control of Network Systems, vol. 5, no. 3, pp. 1322–1334, 2018.
  20. [20]P. E. Paré, F. Liu, C. L. Beck, B. E. Kirwan, and T. Başar, “Analysis, estimation, and validation of discrete-time epidemic processes,” IEEE Transactions on Control Systems Technology, vol. 28, no. 1, pp. 79–93, 2020.
  21. [21]M. Ogura and V. M. Preciado, “Stability of spreading processes over time-varying large-scale networks,” IEEE Transactions on Network Science and Engineering, vol. 3, no. 1, pp. 44–57, 2016.DOI
  22. [22]R. Lambiotte, J.-C. Delvenne, and M. Barahona, “Random walks, markov processes and the multiscale modular organization of complex networks,” IEEE Transactions on Network Science and Engineering, vol. 1, no. 2, pp. 76–90, 2015.DOI
  23. [23]A. Beros, M. Chyba, A. Fronville, and F. Mercier, “A morphogenetic cellular automaton,” in 2018 Annual American Control Conference (ACC). IEEE, 2018, pp. 1987–1992.DOI
  24. [24]V. Martinelli, D. Salzano, D. Fiore, and M. di Bernardo, “Multicellular PD control in microbial consortia,” IEEE Control Systems Letters, vol. 7, pp. 2641–2646, 2023.DOI
  25. [25]S. M. Brancato, D. Salzano, D. Fiore, G. Russo, and M. di Bernardo, “Ratiometric control of two microbial populations via a dual chamber bioreactor,” IEEE Control Systems Letters, vol. 8, pp. 1301–1306, 2024.DOI
  26. [26]A. Giusti, M. Coraggio, and M. di Bernardo, “Local convergence of multi-agent systems towards rigid lattices,” IEEE Control Systems Letters, vol. 7, pp. 2869–2874, 2023.DOI
  27. [27]N. R. Raz and M.-R. Akbarzadeh-T, “Target convergence analysis of cancer-inspired swarms for early disease diagnosis and targeted collective therapy,” IEEE Transactions on Neural Networks and Learning Systems, vol. 33, no. 5, pp. 2132–2146, 2021.DOI
  28. [28]A. Pal, P. Haliti, B. Dharmadhikari, W. Qi, and P. Patra, “Manipulating extracellular matrix organizations and parameters to control local cancer invasion,” IEEE/ACM Transactions on Computational Biology and Bioinformatics, vol. 18, no. 6, pp. 2566–2576, 2020.DOI
  29. [29]A. Anderson and K. Rejniak, Single-cell-based models in biology and medicine. Springer Science & Business Media, 2007.
  30. [30]K. Binder, D. M. Ceperley, J.-P. Hansen, M. Kalos, D. Landau, D. Levesque, H. Mueller-Krumbhaar, D. Stauffer, and J.-J. Weis, Monte Carlo methods in statistical physics. Springer Science & Business Media, 2012, vol. 7.
  31. [31]G. Bhanot, “The metropolis algorithm,” Reports on Progress in Physics, vol. 51, no. 3, p. 429, 1988.
  32. [32]S. P. Meyn and R. L. Tweedie, Markov chains and stochastic stability. Springer Science & Business Media, 2012.
  33. [33]K. M. Yamada, J. W. Collins, D. A. Cruz Walma, A. D. Doyle, S. G. Morales, J. Lu, K. Matsumoto, S. S. Nazari, R. Sekiguchi, Y. Shinsato et al., “Extracellular matrix dynamics in cell migration, invasion and tissue morphogenesis,” International journal of experimental pathology, vol. 100, no. 3, pp. 144–152, 2019.DOI

Paper details

vlnc:W_se8zmrb8
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 3

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Engineering

Not recorded for this paper

View statement
Looking for related papers…
Contents