Compound Race Pathology: A Friction-Theoretic Framework for Multi-Scale Disease Progression and Intervention

Paper 8B · Pødenphant Lund, T. (2026k) · Hypothesis paper · Live on Zenodo

Several major chronic and progressive diseases share a clinical signature that single-mechanism explanations have struggled to capture: chronic multi-system dysregulation, poor response to single-target therapy, large inter-individual heterogeneity, and episodic exacerbations that cannot be reduced to any one pathway. This paper proposes that cancer, autoimmune conditions, post-viral fatigue syndromes (ME/CFS and long COVID), and treatment-resistant depression are instances of a shared mechanism it calls compound race pathology: disease produced by coupled multi-scale RACE-architectures, where intervention bounded at any single scale's control coefficient cannot reach the compound mechanism. It is a hypothesis with falsifiable predictions, and an invitation to test it. It is not clinical advice, and it recommends nothing at the individual level.

DOI (concept)	10.5281/zenodo.20059869
Type	Hypothesis paper (theoretical, with framework-prediction synthesis and falsification criteria)
Cite-letter	2026k
Formal anchor	Metabolic Control Analysis (Kacser & Burns 1973; Heinrich & Rapoport 1974)
Author	Tomas Pødenphant Lund [ORCID]

The construct: coupled multi-scale RACE

The framework inherits its substrate-level mechanism from the RACE architecture (parallel evaluation of competing routes, accumulation to a commit-threshold, irreversible commit that deposits an asymmetric hysteretic trace) developed at the FT level in Paper 1. Its extension here is that biological systems implement this architecture at multiple substrate scales simultaneously, coupled and dynamically so: molecular races (substrate molecules for catalytic capacity, transcription factors for promoter binding), cellular races (apoptosis versus survival, effector versus regulatory phenotype), tissue-microenvironment races (immune surveillance versus evasion), organ-system races (autonomic balance, HPA-axis tone), and behavioural race-resolution all run in parallel and couple back through stress, sleep, and behavioural inputs. Each scale has its own time-constant; phenotype is the resultant. The claim is that small perturbations in one scale's substrate-availability, even within conventional reference ranges, can produce substantial phenotypic effects through propagation across coupled scales.

A structural consequence is outcome-cluster aggregation: when multiple distinct substrate-route configurations point at the same clinical endpoint, the observable outcome-cluster (symptom presentation, disease label, treatment-response category) aggregates over those routes without preserving route-identity. Diagnostic homogeneity masks substrate-route heterogeneity, so response heterogeneity within a diagnosed population tracks differences current stratification does not capture. This is the outcome-cluster-versus-route-identity distinction Paper 8 develops for psychiatry, applied here to compound somatic disease.

The mechanism: the MCA summation theorem, extended

Metabolic Control Analysis defines flux-control coefficients for each enzyme in a pathway, and its summation theorem states they sum to one. The implication: where pathway control is distributed across many steps, single-enzyme intervention produces at most a fraction of the flux change proportional to that enzyme's coefficient, however potent at that target, while combinations across steps can approach the sum of their coefficients in the linear approximation.

The framework's extension is that this bound generalises to multi-scale disease progression under explicit assumptions stated rather than assumed to transfer automatically. Heinrich & Rapoport (1974) formulated MCA generally enough to apply to any system with distributed linear control over a conserved scalar; the biochemistry reading is one instance. The extension treats progression-rate as a quasi-stationary scalar over the intervention window, approximately additive in the linear regime (a modelling choice, not a conservation law), with per-scale coefficients defined as the elasticity of progression-rate to that scale's substrate-modification. The central caveat is stated directly: the summation theorem is a steady-state result, whereas compound disease is non-steady-state (flares, remissions, hysteresis), so the MCA-bound is used as a linear-regime, quasi-stationary approximation over an intervention window, not as a theorem over the full hysteretic trajectory.

Where the assumptions fail, the framework specifies modified formalism organised as a four-regime classification: linear (weak coupling, strict summation), non-linear (strong coupling, interaction terms detectable via factorial design), cascade (critical coupling, Buldyrev et al. 2010 interdependent-network percolation, bimodal cascade-event distribution), and saturation (Hill-equation, cooperativity coefficient greater than one). The classification is falsifiable as a unit: the four regimes are claimed jointly exhaustive and individually discriminable by distinct signatures, so a compound disease whose well-resolved combination-effect pattern matches none would refute the extension rather than license a fifth regime. The framework does not yet provide measured control coefficients for any specific disease; it names rheumatoid arthritis as the most tractable starting point, because O'Dell triple-DMARD effect-size data (O'Dell et al. 1996, 2002, 2013) is dense enough to back-fit coefficients to existing trial data, and HFrEF as a further candidate where four-drug-class data is densest.

The four pathology classes (and one out-of-set probe)

• Carcinogenesis — the paradigmatic compound race: no single molecular event produces the cancer phenotype, which requires accumulation across years to decades (Hanahan & Weinberg 2000, 2011; Vogelstein et al. 2013). The DNA-damage-versus-repair race runs continuously against 10⁴–10⁵ damage events per cell per day (Lindahl 1993); apoptosis-versus-survival and immune-surveillance-versus-evasion races are won through compound dysregulation. Multi-factor lifestyle prevention is MCA-consistent: EPIC-Potsdam (Ford et al. 2009) reports 36% cancer reduction at four-factor adherence.
• Autoimmune disease as compound-tipping — flares emerge when multiple individually-mild contributions cross threshold approximately simultaneously; no single trigger reliably initiates exacerbation (Costenbader & Karlson 2006; Parks et al. 2014). Combination therapy systematically outperforms monotherapy: triple-DMARD in RA achieved 50% improvement in 77% versus 33% on methotrexate-monotherapy (O'Dell et al. 1996), durable at eleven-year follow-up (FIN-RACo, Mottonen et al. 1999); SONIC showed combination biologic-plus-immunomodulator at 56.8% versus 44.4% versus 30.0% steroid-free remission (Colombel et al. 2010). Within-individual remission-versus-flare cycling is read as a bimodal hysteresis-attractor signature.
• ME/CFS and long COVID — persistent multi-system dysregulation after viral infection, no consistent single biomarker, no reliable single-target therapy (Komaroff & Lipkin 2021; Davis et al. 2021). Read as compound failure-to-re-equilibrate across coupled scales (mitochondrial, autonomic, immune, neuroendocrine, central-sensory) held in a dysregulated attractor by mutual coupling. Post-exertional malaise is re-interpreted as hysteresis-overshoot: exertion pushes demand past a commit-threshold, the system commits to an energy-deficit attractor, and the delayed-and-prolonged time-course is the signature of state-commitment rather than linear depletion.
• Treatment-resistant depression — affects roughly 30% of major-depression patients and is mechanistically distinct, not simply more severe (Trivedi et al. 2006; Rush et al. 2006). Treated as a somatic-coupled compound mechanism: in a substantial subgroup the phenotype is maintained by coupled dysregulation across inflammatory, HPA-axis, metabolic, and microbiome scales, which is why single-target monoamine modulation under-addresses it. Consistent with the Raison et al. 2013 infliximab RCT: no main effect, but a significant interaction with baseline CRP (high-CRP improved, low-CRP worsened), read as substrate-vector-dependent response.
• HFrEF (out-of-set applicability probe) — chosen deliberately from outside the framework's construction-set. Its four-drug-class standard of care (ARNI/ACE-inhibitor, beta-blocker, mineralocorticoid-receptor antagonist, SGLT2-inhibitor; synthesis in Vaduganathan et al. 2020) is among the best-quantified combinations in medicine. The framework reads it as a cascade-plus-saturation case, matching the empirical diminishing-marginal-returns pattern. The paper is explicit that this shows applicability, not confirmation, and registers forward COPD and T2DM regime-assignments (fixed before the data) as the genuinely adversarial tests.

Intervention implications

The multi-target principle. Because disease-progression control is distributed across multiple races at multiple scales, single-target intervention is bounded (in the linear regime) by the targeted scale's control coefficient, and only multi-target intervention can approach the sum across contributing scales. This is a principled re-statement of the empirical combination-therapy advantage, with the specification that the achievable advantage is governed by the distribution of coefficients and the operative regime. The MCA-bound is the sharpest claim: potency at a target does not translate into progression-control if that target's coefficient is low, which is offered as the reason highly-potent single-target drugs often produce disappointing progression effects.

Prevention. Pre-clinical intervention acts on a substrate where the disease-attractor is not yet hysteretically locked, so the same multi-factor intervention is predicted to produce disproportionate benefit applied pre-threshold versus post-threshold. This is grounded in the additive-only asymmetry (deposition runs faster than shallowing) that Paper 8D and the molecular-hysteresis thread develop, which makes prevention mechanically privileged over reversal.

Substrate-vector stratification. The operational diagnostic is to stratify patients by a multi-axis substrate-vector aligned to the contributing scales (molecular-metabolic, cellular, tissue-inflammatory, organ-system-autonomic, neuroendocrine, behavioural) rather than by outcome-cluster label. Its framework-distinctive, falsifiable addition beyond standard precision-medicine biomarker stratification is that intervention at one axis is predicted to propagate to coupled axes on a delayed (weeks-to-months) timescale via cross-scale commit-pressure, a trajectory distinct from fast, dose-dependent pharmacological pleiotropy. Paper 8C develops the six-axis biomarker panel as concrete starting infrastructure.

Hysteresis-fighting interventions. Where standard intervention modulates an ongoing race's commit-pressure, hysteresis-fighting intervention targets the deposited trace that holds a system in a locked attractor. The framework distinguishes three categories. Category A is cellular substrate-replacement, sub-classified by reversibility (A1 cellular-reversible, with the CASTLE 2026 SLE/IIM/SSc CAR-T basket trial read as a substrate-reversibility hypothesis; A2 mixed; A3 fibrotic halt-only). Category B is substrate-modification with a consolidation requirement, exemplified by the Wilkinson 2017 ketamine-plus-psychotherapy paradigm (ketamine opens a plasticity window, psychotherapy consolidates a new attractor, producing durability that ketamine alone, median time-to-relapse 13–18 days, does not). Category C is state-perturbation requiring continuous maintenance, with ECT (84% relapse within six months without continuation; Sackeim et al. 2001) and sleep-deprivation response (83% relapse after one night of recovery sleep) as paradigms. The taxonomy predicts post-withdrawal durability from the intervention's substrate-mechanism, which standard pharmacological classification does not.

Predictions and falsification criteria

The paper states ten predictions (P8b.1–P8b.10), each with framework-basis and operationalisation, and a separate R1–R10 negative-prediction set. R3 and R5 are framework-pivotal: their joint disconfirmation forces framework-rejection, not revision, with no graceful-degradation clause; the component-tests force only component-revision. Several predictions are testable on existing public-access data.

• P8b.1 — combination-effect MCA-bound. In the linear regime, combination-effect approaches the sum of control coefficients across modified scales; operationalised by factorial-design back-fitting to existing RA triple-DMARD trial data (O'Dell et al. 1996, 2002, 2013).
• P8b.6 — substrate-vector response-stratification in TRD. Multi-target-augmentation response is stratified by substrate-vector; the inflammatory-axis-high subgroup responds to anti-inflammatory adjunct. Operationalised by re-analysing existing biomarker-stratified trial data (the Raison et al. 2013 infliximab RCT by baseline CRP).
• P8b.7 / P8b.10 — hysteresis-durability and response-shape. Hysteresis-fighting interventions show bimodal (attractor-shift) rather than continuous response, with a durability gradient A1 > A2 > A3, and Category-B (ketamine-plus-CBT) durable post-withdrawal versus Category-C (ECT-without-continuation) relapse.
• P8b.9 — Loewe-within-scale plus Bliss-across-scale. Combination-drug effects decompose into Loewe-additive within-scale plus Bliss-additive across-scale components, resolving the “synergistic by Bliss, antagonistic by Loewe” pharmacology paradox; operationalised via drug-target-mapping (DrugBank, Reactome) plus SynergyFinder-style scoring.
• R3 (pivotal) — substrate-vector predicts response. Falsified if substrate-vector stratification adds no incremental AUC over a diagnosis-cluster baseline in three independent multi-axis TRD cohort re-analyses by 2030.
• R5 (pivotal, deferred-falsification) — substrate-reversibility gradient. Predicts SSc CAR-T drug-free remission below SLE (A1 > A3) in an n ≥ 100 follow-up trial; gradient-flattening or reversal refutes the hysteresis-substrate-coupling mechanism. The clock starts when an adequately-powered SSc CAR-T trial reports, which does not yet exist.

What this paper is, and is not

What it is: a unifying mechanistic hypothesis that four separately-studied pathology classes are instances of one compound-race mechanism; an extension of Metabolic Control Analysis to multi-scale disease under explicitly stated assumptions, with a four-regime classification specifying where the linear approximation holds; a multi-target intervention principle, a substrate-vector-stratification diagnostic, and a three-category hysteresis-fighting taxonomy; and ten predictions plus an R1–R10 falsification set, several runnable on existing trial and cohort data.

What it is not: it is not a clinical guideline and makes no patient-level recommendations; it proposes no intervention-protocol for any of the four classes and specifies no doses. It is a thinking-tool for clinical researchers, not a decision-algorithm. It is not a unification of all complex chronic disease: acute trauma, monogenic disorders, and single-pathogen infectious disease are explicitly out of scope. Its scope is multi-scale chronic disease progression where compound mechanism is empirically documented and single-target therapy is systematically under-powered. It is hypothesis-plus-invitation, offered for domain experts to test, extend, falsify, or refine.

Limitations (stated plainly in the manuscript)

The largest limitation is structural: the author is an independent researcher without a clinical collaborator currently in place, without institutional cohort-access, and without domain credentials in oncology, immunology, post-viral medicine, or treatment-resistant-depression neurobiology, so the predictions are specified for testing by experts who have the substrate-access this paper does not. No new data is collected: the paper unifies existing meta-analytic and trial findings, so its empirical claims are framework-predictions, not results. The control coefficients are predicted, not measured for any specific compound disease, so until they are the MCA-bound is structural rather than numerical. The biological claims are framed as predictions: where the framework re-interprets established findings (multi-step carcinogenesis, the combination-therapy advantage), the cited biology is real and current while the synthesis of it is the hypothesis. The mathematics is a deliberate simplification: the MCA-extension treats progression-rate as a quasi-stationary additive scalar in the linear regime, with cascade, saturation, and non-linear coupling as the biologically-realistic regimes whose quantitative predictions are correspondingly less developed. The R1–R10 set is specified precisely so the framework can be disconfirmed.

Connections to other papers in the series

• Paper 1 (Friction Theory) — the substrate-universal framework: RACE architecture, commit-pressure, hysteresis, and the FT-level adoption of Metabolic Control Analysis that this paper operationalises for multi-scale disease.
• Paper 8 (Pressure, Hysteresis, and Experience) — the parent psychiatric clinical-intervention paper, supplying the outcome-cluster-versus-route-identity distinction, the intervention taxonomy, and the additive-only prevention asymmetry. Scope-boundary: Paper 8 treats TRD as a psychiatric-clinical phenomenon; Paper 8B treats somatic compound disease, with TRD as a compound-multi-scale extension.
• Paper 8C (Translational Research-Program) — the trial-design companion; its six-axis biomarker panel is the concrete operationalisation of this paper's substrate-vector-stratification requirement.
• Paper 8D (Treating the Base, Not the Top) — the sub-threshold cofactor-support paper; its compound-profile reasoning is the sub-threshold counterpart to this paper's multi-target argument, and it shares the prevention-privilege asymmetry.
• Learning — the learning-and-memory thread, where the hysteresis and consolidation machinery this paper borrows (attractor-locking, the Category-B durability prediction) is developed.

Read the paper

The full paper is on Zenodo (concept DOI 10.5281/zenodo.20059869):

Read on Zenodo → · Plain English version · Dansk version

This page summarises a hypothesis paper. It is not medical advice, not a diagnostic manual, and does not tell any individual what to do. The biological claims are predictions to be tested, not proven facts. Any decision about treatment belongs with a qualified clinician.