Source-linked AI summary
Exploiting ecological principles to better understand cancer progression and treatment
David Basanta, Alexander R. A. Anderson
TL;DR
Cancer progression is framed as disruption of homeostasis emerging from interactions among tumour cells and their environments. The paper uses ecological perspectives and mathematical models to discuss tumour dynamics and treatment, arguing that this view complements cancer research and may inform improved therapies.
Problem
Cancer progression involves interactions among tumour cells and their environments, making it insufficient to view cancer only as a genetic process.
Source: §1 · pp. 1, 3Method
The paper applies an ecological perspective and mathematical models of tumour phenotypes and microenvironmental interactions to examine cancer progression.
Source: AG INV GLY · pp. 4–5, §3 · p. 9Results
The authors conclude that an ecological view complements existing cancer research and may lead to better understanding of cancer biology and improved therapies.
Source: §3 · p. 10Takeaways & Limitations
Representing ecosystem interactions with weights can help investigate how tumours disrupt homeostasis, develop, and respond to perturbations.
Source: §3 · p. 9Most EGT models do not explicitly represent space, focus on relative rather than absolute population sizes, and often use abstract parameters that prevent meaningful extrapolation to population size.
Source: Interactions and cooperation · p. 7
Abstract
from arXiv · showhide
A small but growing number of people are finding interesting parallels between ecosystems as studied by ecologists (think of a Savanna or the Amazon rain forest or a Coral reef) and tumours1-3. The idea of viewing cancer from an ecological perspective has many implications but fundamentally, it means that we should not see cancer just as a group of mutated cells. A more useful definition of cancer is to consider it a disruption in the complex balance of many interacting cellular and microenvironmental elements in a specific organ. This perspective means that organs undergoing carcinogenesis should be seen as sophisticated ecosystems in homeostasis that cancer cells can disrupt. It also makes cancer seem even more complex but may ultimately provides isights that make it more treatable. Here we discuss how ecological principles can be used to better understand cancer progression and treatment, using several mathematical and computational models to illustrate our argument.
1. Cancer and ecosystems
Cancer ecology treats cancer as disruption of a healthy organ ecosystem, where cellular and environmental interactions shape tumour evolution. Mathematical and computational models can help study this complexity.
1. Cancer and ecosystems
- Cancer ecology frames tumorigenesis as disruption of tissue homeostasis through changes in the microenvironment, invading species, or local mutation.Healthy tissues balance cellular and environmental components; at its simplest, tissue homeostasis balances proliferation and apoptosis.Source: §1 · pp. 1–3
- Tumour cells that adapt to a new environment may invade and metastasise, with their prospects shaped by both that environment and the cellular species of the original site.The seed-soil hypothesis holds that the metastatic site matters as much as the metastatic cells [5]; cancer evolution has also been emphasized [1,3].Source: §1 · p. 2
- Homeostasis is maintained through cellular and environmental mechanisms, including adhesion, metabolic factors, growth factors, and stroma, with feedback between them.Mutant cells must significantly modify baseline phenotypes and ignore environmental signals to escape homeostatic control.Source: §1 · p. 2
- Tumours can show distinct genetic profiles but similar phenotypes, a pattern reported across cancers including lung, kidney, prostate, and brain [12–14].The authors suggest that dialogue between tumour cells and their environments selects clones best adapted phenotypically to survive.Source: §1 · pp. 2–3
- Evolutionary game theory can examine how tumour phenotypes interact and how those interactions may shape cancer progression [15].The Hawk-Dove example contrasts aggressive and meek strategies; a tumour analogue contrasts motile and proliferative phenotypes.Source: §1 · p. 3
- Because tumour ecosystems involve heterogeneous cells, healthy cells, and the physical microenvironment, purely experimental approaches are described as impractical for understanding their complexity.The authors point to existing mathematical and computational tools for studying ecosystems across scales and complexity.Source: §1 · pp. 3–4
2. Mathematical tools
Evolutionary game theory models how a strategy’s outcome depends on the strategies of other players. Applied to cancer, it cautions that indiscriminate cell removal may produce only a temporary effect.
2. Mathematical tools
- Evolutionary game theory models how strategies fare through interactions with other players, with natural selection favoring long-term reproduction.The authors describe simple EGT models for understanding homeostasis and its disruption in cancer.Source: §2 · p. 4
- Indiscriminately removing most but not all cancer cells may have only a temporary effect, as tumour cell numbers can eventually return to and exceed their original level.The passage says many EGT models propose an alternative based on changing the way cells interact, but does not specify that approach here.Source: §2 · p. 4
AG INV GLY
Tumour progression reflects interactions among cancer cells and their microenvironment, not genetic change alone. Ecological and mathematical models show how these interactions shape tumour phenotypes and treatment responses.
AG INV GLY
- Adding glycolytic phenotypes increases motile cells’ relative fitness advantage, producing more aggressive tumours than the original two-phenotype game.The intermediate glycolytic phenotype need not be evolutionarily successful in the long term for heterogeneity to alter the game’s dynamics.Source: AG INV GLY · pp. 4–5
- Evolutionary treatment strategies include selecting for easier-to-treat phenotypes and sequencing therapies so that evolution directs tumours toward greater susceptibility.For lung cancer, chemotherapy before a p53 vaccine produced better responses than either one treatment alone or the reversed sequence [20,21]; an EGT model considered three treatment-sensitivity populations [22].Source: AG INV GLY · pp. 5–6
- Treatment that kills cooperating stromal cells selects for the I tumour population, leaving a smaller tumour with unaffected growth potential that is completely resistant.The example begins with a tumour described as highly susceptible to the treatment.Source: Figure 2 · p. 5, AG INV GLY · p. 5
- Cooperation can emerge under selection when interacting parties benefit, and tumour cells may both compete for resources and cooperate by producing growth factors [2,23,24].EGT can examine how these interactions affect tumour evolution and movement toward or away from homeostasis.Source: Interactions and cooperation · pp. 6–7
- Most EGT models omit explicit spatial interactions and focus on relative subpopulation proportions, limiting direct inference about absolute population sizes.Absolute numbers may be inferred when initial populations are known and the model is properly parameterized, but abstract parameters often prevent meaningful extrapolation.Source: Interactions and cooperation · p. 7
- Individual-based models represent cells individually and capture space and time; hybrid models combine discrete cells with continuous environmental fields to study multiscale interactions.A prostate cancer model used stromal–tumour interactions via TGFβ and data from in vivo and in vitro experiments to study emergent properties over timescales spanning decades.Source: Interactions and cooperation · pp. 7–8
- Different microenvironments produce distinct tumour morphologies; harsher settings select for more aggressive phenotypes, while more favorable settings can yield more heterogeneous tumours.In a parameterized HDC model, aggressive cells were predicted to become essentially microenvironmentally independent, producing their own niche as required.Source: Interactions and cooperation · p. 8
- Combining models can connect evolutionary cell interactions with spatial effects and link cellular behavior to pathway networks [34,35].The cited examples combine EGT with a cellular-automaton individual-based model, or networks with an individual-based model.Source: Interactions and cooperation · pp. 8–9
3. Discussion
The discussion frames cancer progression as an ecosystem-level process shaped by interactions among cellular populations and the tumour microenvironment. It argues that ecological models can complement gene-focused research, while emerging microscopy may support more holistic, quantitative models.
3. Discussion
- Ecological models are phenotype-centered and can miss biology below that scale, while gene-centric research has rich molecular data; the approaches can complement one another.The discussion notes that integrated experimental study is difficult without theoretical approaches, including agent-based models that bridge genotypic and phenotypic scales [36,37].Source: §3 · p. 9
- Cancer progression emerges from interactions among multiple cellular species and the tumour microenvironment, with metastasis depending on the suitability of a secondary site for survival and colonisation.Suitability includes resemblance to key features of the primary site and access to nutrients and room for growth.Source: §3 · p. 9
- A prostate-to-bone metastasis ecosystem includes tumour cells, bone-cell populations, and stem cells whose interactions involve competition and cooperation for resources.The figure is a simplification of the ecosystem.Source: §3 · p. 9, Figure 6 · p. 10
- High-throughput automated microscopy is making substantial cellular information more accessible, which may support holistic, quantitative models whose parameters and predictions can be compared with experiments.The paper calls such approaches “qolistic” [41].Source: §3 · p. 10
- The ecological view is presented as a complement to decades of cancer research, with the prospect of improving biological understanding and therapies.Source: §3 · p. 10