Endogenous versus Exogenous Origins of Diseases

Many illnesses are associated with an alteration of the immune system homeostasis due to any combination of factors, including exogenous bacterial insult, endogenous breakdown (e.g., development of a disease that results in immuno suppression), or an exogenous hit like surgery that simultaneously alters immune responsiveness and provides access to bacteria, or genetic disorder. We conjecture that, as a consequence of the co-evolution of the immune system of individuals with the ecology of pathogens, the homeostasis of the immune system requires the influx of pathogens. This allows the immune system to keep the ever present pathogens under control and to react and adjust fast to bursts of infections. We construct the simplest and most general system of rate equations which describes the dynamics of five compartments: healthy cells, altered cells, adaptive and innate immune cells, and pathogens. We study four regimes obtained with or without auto-immune disorder and with or without spontaneous proliferation of infected cells. Over all regimes, we find that seven different states are naturally described by the model: (i) strong healthy immune system, (ii) healthy organism with evanescent immune cells, (iii) chronic infections, (iv) strong infections, (v) cancer, (vi) critically ill state and (vii) death. The analysis of stability conditions demonstrates that these seven states depend on the balance between the robustness of the immune system and the influx of pathogens.

💡 Research Summary

The paper tackles a fundamental question in immunology and disease biology: why does the immune system appear to require a continual influx of pathogens to maintain homeostasis? Building on the evolutionary premise that host immunity co‑evolved with a pathogen-rich environment, the authors propose that a basal level of microbial exposure is essential for the immune system to stay “primed” and to rapidly adjust to infection bursts. To formalize this idea, they construct the simplest possible set of ordinary differential equations (ODEs) that capture the dynamics of five interacting compartments: (1) healthy host cells (H), (2) altered or damaged cells (A) which may represent early‑stage cancer or infected cells, (3) innate immune cells (I) such as macrophages and NK cells, (4) adaptive immune cells (C) such as T‑ and B‑lymphocytes, and (5) external pathogens (P).

Each compartment is linked by biologically motivated terms: healthy cells die naturally and are lost to infection; altered cells arise from infection of healthy cells and may proliferate autonomously at rate α; innate and adaptive cells are activated by pathogen and altered‑cell signals, but also undergo natural turnover and can be driven into auto‑reactive behavior at rate β. Pathogen dynamics are driven by an external influx λ (e.g., environmental exposure, surgery, or medical procedures) and cleared by both innate and adaptive arms. The model therefore contains a small set of key parameters: pathogen influx λ, immune activation and decay rates, autonomous proliferation α of altered cells, and auto‑immunity strength β.

Four distinct regimes are examined by toggling two binary features: the presence or absence of auto‑immunity (β = 0 vs. β > 0) and the presence or absence of spontaneous proliferation of altered cells (α = 0 vs. α > 0). For each regime the authors solve for steady‑states (fixed points) of the ODE system and assess linear stability via the Jacobian eigenvalues. Across all regimes, seven qualitatively different equilibria emerge naturally from the mathematics:

1. Robust healthy state – high H, I, C; low P; immune system efficiently controls pathogens.
2. Immune‑evanescent healthy state – H remains high but I and C collapse to near zero; the host is vulnerable to sudden pathogen surges.
3. Chronic infection – moderate, persistent levels of P and A coexist with a steady, but not maximal, immune response.
4. Acute/strong infection – P spikes, I and C surge dramatically, then may overshoot and collapse, leading to system instability.
5. Cancer‑like state – A dominates while immune compartments are suppressed; reflects unchecked proliferation of altered cells.
6. Critically ill (auto‑immune) state – excessive β drives I and C to hyper‑activation, causing collateral damage to healthy tissue and eventual collapse.
7. Death – all compartments either vanish or P diverges, representing physiological failure.

The location of each equilibrium in parameter space is governed primarily by the balance between immune robustness (strength of I and C activation, low decay rates) and pathogen influx λ. Low λ combined with strong immunity yields the robust healthy state; moderate λ can sustain the immune‑evanescent or chronic infection states; high λ or weakened immunity pushes the system toward acute infection, cancer, or critical illness. Notably, the model predicts that completely eliminating pathogen exposure (λ → 0) leads to immune cell extinction, supporting the authors’ conjecture that some microbial “noise” is essential for immune vigilance.

Sensitivity analyses reveal clear therapeutic implications. Reducing λ (e.g., through improved hygiene, prophylactic antibiotics, or peri‑operative infection control) can shift the system from a chronic or acute infection basin into the healthy basin, provided immune competence is maintained. Conversely, boosting immune activation (e.g., vaccines, checkpoint inhibitors) expands the robust healthy region but may also enlarge the auto‑immune basin if β is inadvertently increased. The presence of autonomous altered‑cell proliferation (α > 0) dramatically enlarges the cancer‑like basin, suggesting that early detection and targeted suppression of such proliferation are crucial.

The authors acknowledge several limitations. The model aggregates heterogeneous cell types into five coarse compartments, ignores spatial heterogeneity, and treats parameters as constant rather than dynamically regulated. Parameter estimation from clinical data remains an open challenge. Nonetheless, the framework offers a unifying, mathematically tractable lens through which diverse disease phenotypes—infectious, neoplastic, and immune‑mediated—can be interpreted as alternative attractors of the same underlying dynamical system. Future work is proposed to incorporate tissue‑specific modules, stochastic pathogen exposure, and patient‑specific parameter fitting, paving the way for personalized predictions of disease trajectories and optimal intervention strategies.