The Apoptotic Budget

Abstract

Parkinson’s disease patients have 18–50% lower cancer rates. Cancer patients show reduced neurodegeneration. This inverse correlation has been confirmed across 17 million participants and dozens of meta-analyses, yet no structural explanation exists. The shared pathways are well-catalogued — p53, PINK1/Parkin, mTOR, mitochondrial dysfunction — but they are treated as parallel coincidences, not as competing draws on a finite system.

We used the OMUO Genesis Engine, a geometric knowledge discovery system that encodes domain concepts as algebraic vectors constrained to an E8 lattice, to run four synthesis experiments across this problem space. The engine processed over 120 molecular actors, generated 600+ structural bridges, and converged on a central finding: the body enforces a fixed total apoptotic capacity. A system chronically consuming that capacity on neuronal death has no budget left for tumour suppression. Cancer and neurodegeneration are not opposite diseases. They are opposite failures of the same resource allocation system.

The engine further identified VDAC1 — the voltage-dependent anion channel on the mitochondrial outer membrane — as the physical chokepoint where this trade-off plays out. Three different misfolded proteins (alpha-synuclein, amyloid-beta, mutant SOD1) jam this gate by shape, not by chemistry. Cancer cells exploit the same gate by parking hexokinase there. We propose that VDAC1 occupancy constitutes a competitive docking geometry problem, and that this framing unifies drug target discovery for both diseases.

Keywords: Parkinson’s disease, cancer, inverse correlation, apoptosis, VDAC1, geometric synthesis, drug targets, E8 lattice, mitochondrial gatekeeper, competitive docking

1. Introduction

Here is the puzzle. A person with Parkinson’s disease is losing neurons. The cells in their substantia nigra are dying too fast, too early. And yet that same person is significantly less likely to develop cancer — a disease defined by cells refusing to die. A meta-analysis by Bajaj et al. covering 107,598 PD patients found a 27% decreased risk of all cancers [1]. A larger study across 17 million participants confirmed this inverse association holds for both smoking-related and non-smoking-related cancers [2]. The most recent data, a 2025 systematic review of colorectal cancer risk in PD, reported the same pattern and pointed to enhanced pro-apoptotic signalling as a likely driver [3].

The problem is not the evidence. The evidence is overwhelming. The problem is the explanation.

The published literature identifies shared molecular pathways — p53-MDM2 feedback, PINK1/Parkin mitophagy, mTOR/autophagy, NF-κB inflammatory signalling, ferroptosis, gut-brain axis involvement [4, 5, 6]. But these pathways are catalogued side by side, like suspects in a lineup, with no unifying geometry. Nobody has asked the structural question: if the same pathways produce opposite diseases in opposite directions, is there a single finite resource being allocated between them?

We asked that question. The answer surprised us.

2. Methods

2.1 The Genesis Engine

The OMUO Genesis Engine is a geometric knowledge discovery system. It encodes domain concepts as complex phasor vectors constrained to an E8 lattice — a 240-direction mathematical structure with known symmetry properties. When two concepts are algebraically combined (bound), the resulting vector occupies a new position in the lattice. The bridge name and structural law are generated by a language model, but the geometry is deterministic: same inputs always produce the same coordinates, distances, and lattice positions.

The engine is not a language model performing analogy. It is an algebraic system that finds what emerges when two concepts combine — structurally, not semantically. Vector databases find things similar to X. The Genesis Engine finds what happens when X and Y are algebraically combined. This is the difference between similarity search and structural synthesis.

Quality gates filter every bridge: BST (binding stability test) depth measures how many oscillation cycles a bridge survives before settling; Soares Resonance measures geometric alignment with fundamental constants. Only bridges that pass both filters are retained. The system is described in detail in prior publications [7, 8].

2.2 Experimental Design

We conducted four synthesis runs, each approaching the Parkinson’s–cancer axis from a different angle:

Combined, the four runs produced over 600 structural bridges. We retained only those meeting quality thresholds and report the strongest findings below.

3. Results

3.1 The Apoptotic Budget: A Finite System-Wide Allocation

The engine’s highest-confidence finding, appearing independently across all four runs: the body does not choose between cell death and cell survival. It enforces a fixed total apoptotic capacity. When that capacity is chronically consumed by one process — say, killing damaged neurons — there is no quota left for the other.

The strongest bridge in the entire study was “Cellular Attention Span” (Soares Resonance 1.962), which emerged from binding the Apoptotic Budget concept with a quality-control overdrive pattern. The law: both systems treat cellular focus as a finite resource, where hyper-vigilance on one quality-control task (checking for misfolded proteins) necessarily drains the budget for executing another (programmed cell death). A cell can be too attentive to live properly.

The second-strongest bridge, “Energy Commitment Threshold” (SR 1.965), revealed that the cell’s survival switch and its fuel-selection mechanism draw from the same limited pool. The cell is either betting its energy on living longer or on building a specific protein machine. Both bets drain the same decisive energy.

This is a genuinely different framing from the published literature. The standard approach identifies shared pathways — p53 appears in both diseases, PINK1/Parkin appears in both diseases — and catalogues them. The budget framing says something stronger: these pathways are not just shared, they are in zero-sum competition. A neuron that is chronically running its apoptotic machinery at low volume is physically occupying the slots, consuming the molecular tokens, and draining the processing bandwidth that would otherwise be available for tumour suppression.

We want to be honest about what this means. The engine did not discover new proteins. It discovered a structural relationship between known proteins that reframes the question. And the reframe has a specific, testable prediction: the molecular machinery of apoptosis has finite throughput, and diseases compete for that throughput.

3.2 VDAC1: The Physical Chokepoint

Run 3 focused entirely on VDAC1, the voltage-dependent anion channel that serves as the primary gate on the mitochondrial outer membrane. It is the most abundant protein on the mitochondrial surface, interacts with over 100 different proteins, and controls the flow of ATP, ADP, NADH, calcium, and metabolites between mitochondria and cytosol [9, 10].

The engine found something that, as far as we can determine, nobody has said clearly before: VDAC1 occupancy is a competitive docking geometry problem, and this single framing unifies the cancer and neurodegeneration literatures.

Three different misfolded proteins — alpha-synuclein in Parkinson’s, amyloid-beta in Alzheimer’s, mutant SOD1 in ALS — all jam the same mitochondrial gate [10, 11]. They share no sequence homology. They converge on VDAC1 by presenting a non-native structural interface the channel was not evolved to reject. The bridge “Channel Jamming by Shape” captured this precisely.

Meanwhile, cancer cells exploit the same slot by parking hexokinase on VDAC1, which gives them preferential access to mitochondrial ATP while simultaneously blocking the apoptotic pathway [9, 12]. The bridge “Occupancy Hijack” described this as a single, high-affinity tenant in a critical gatekeeper slot rewiring the entire system’s fate.

But here is the finding that excited us most. The bridge “Ring-Clamp Geometry” revealed that VDAC1 oligomerisation — where multiple VDAC1 proteins assemble into a larger ring — kills cancer cells by opening the ring too wide (releasing cytochrome c to trigger apoptosis) and kills neurons by jamming the ring shut (blocking metabolite exchange). Cell fate is set by the mechanical tolerance of a single molecular gasket.

Recent published work supports this direction. A January 2026 paper in Nature Communications identified NCATS-SM0225, a small molecule that binds all three VDAC isoforms and selectively kills cancer cells through calcium disruption [13]. A February 2026 paper in The FEBS Journal reported new VDAC antagonist molecules that bind a druggable NADH pocket and alter channel selectivity [14]. And a December 2024 review in Biomolecules explicitly called VDAC1 a key player in the mitochondrial landscape of neurodegeneration and highlighted its interactions with alpha-synuclein, amyloid-beta, and SOD1 [10].

What none of these papers do is frame the cancer and neurodegeneration roles as a single geometric competition for the same physical docking space. They treat VDAC1-in-cancer and VDAC1-in-neurodegeneration as separate problems. Our engine unified them.

3.3 Key Bridges: Summary Table

Table 1. Highest-confidence bridges across four synthesis runs. SR = Soares Resonance. Only bridges with SR > 1.7 shown.

3.4 The VDAC1 Docking Geometry

Table 2. Key VDAC1 bridges from Run 3.

3.5 Drug Target Implications

1. Threshold Tuning, Not Blocking. The bridge “Threshold Tuning Knob” predicts that the ideal drug at the Parkinson’s–cancer intersection would not block death or promote survival. It would recalibrate the cell’s sensitivity to its own internal signals — making it neither deaf nor hypersensitive.

2. Voltage-Gated Selective Eviction. VDAC1’s voltage dependence creates a natural timing mechanism. The “Voltage-Gated Occupancy Filter” bridge suggests it may be possible to exploit the cell’s own membrane potential to selectively evict misfolded protein aggregates from VDAC1 without displacing protective hexokinase binding.

3. Senolytic Collateral Damage. The “Signal Clearance Bottleneck” bridge predicts that dasatinib + quercetin senolytic therapy, by clearing senescent cells, may create a fatal backup in neuronal survival signalling. Both processes compete for the same clearance machinery. This is a testable and clinically relevant warning.

4. Semaglutide as Resource Rationing Override. The bridge “Demand-Sensing Priority Override” predicts that semaglutide (GLP-1 receptor agonist) works not by targeting Parkinson’s specifically, but by hijacking the body’s master system for deciding which cells get to consume resources first. A February 2025 Danish nationwide cohort study confirmed that GLP-1 receptor agonists significantly reduce Parkinson’s risk [15]. Preclinical data shows semaglutide reduces alpha-synuclein levels in MPTP mouse models [16]. Our geometric framework suggests these effects arise because semaglutide recalibrates the metabolic priority queue, shifting the apoptotic budget allocation.

5. Slot Affinity Drift as Ageing Biomarker. The prediction that ageing changes the VDAC1 lock itself — not just which keys are available — maps onto known age-related changes in VDAC1 post-translational modifications (phosphorylation by GSK3-β, deacetylation by SIRT3). Measuring VDAC1 occupancy ratios as a function of age could yield a novel biomarker for Parkinson’s and cancer risk.

4. Discussion

4.1 What Is Genuinely Novel

We want to be precise about what the engine found and what was already known.

The inverse correlation between Parkinson’s and cancer is well-established. The shared molecular pathways are well-catalogued. The role of VDAC1 in both diseases has been individually studied. None of this is new.

What is new:

The budget framing. The published literature identifies shared pathways but treats them as coincidental overlap. We propose they are zero-sum competitors for a finite allocation of apoptotic capacity. This reframes the question from “what do these diseases share?” to “what are they competing for?”

VDAC1 as competitive docking geometry. Published work treats VDAC1’s role in cancer (hexokinase binding, Warburg metabolism) and its role in neurodegeneration (misfolded protein blockade) as separate problems. We unified them as one geometric competition for the same physical docking space.

Death as bandwidth-limited, not threshold-limited. The “Finality Throughput Governor” bridge predicts that a cell’s commitment to death is gated not by a chemical threshold but by the clearance rate of the signals that authorise it. You can stall a point of no return by clogging its exit paperwork.

Semaglutide’s mechanism as resource rationing override. Published work on GLP-1 agonist neuroprotection focuses on insulin signalling and neuroinflammation. The engine’s framing — that semaglutide works by recalibrating the metabolic priority queue — offers a complementary and potentially unifying explanation.

4.2 Limitations

We need to be straightforward about what this work cannot do.

The Genesis Engine finds structural patterns in concept space. It does not run wet-lab experiments. The “apoptotic budget” is a geometric discovery — a structural relationship that the engine identified as the most stable and resonant explanation for the data. It is not, and cannot be, experimental proof that such a budget exists as a measurable physiological quantity. The bridge names and laws are generated by a language model interpreting geometric positions. The geometry is deterministic; the interpretation is not.

We estimate that roughly 30–40% of the high-quality bridges in these runs are genuinely non-obvious to domain experts, 40–50% are correct but already known (“yes, we knew that, but we wouldn’t have phrased it that way”), and 10–20% are noise. The later cycles of each run — beyond approximately 50 bridges — tend toward repetitive variations on the same theme (signal congestion, bandwidth saturation), and we have aggressively filtered these from the reported results.

And one more thing. The engine is built by one person with no physics or mathematics degree, using AI collaboration. We state this openly because it is relevant to how the work should be evaluated. The geometry is deterministic and reproducible. The interpretation is ours. The testable predictions stand or fall on experimental validation, not on credentials.

4.3 Connection to Emerging Research

Several recent developments align with the engine’s findings. The VDAC1 drug-discovery space is accelerating: new antagonist molecules have been identified that compete with NADH for a shared binding pocket on VDAC1 [14], and a novel small-molecule ERAD inhibitor was shown to bind all three VDAC isoforms with selectivity for cancer cells over healthy tissue [13]. The “competitive docking” framing our engine proposed maps directly onto these pharmacological advances.

The GLP-1 agonist story is moving fast. A phase 2 randomised trial showed lixisenatide prevented worsening of PD motor symptoms over 12 months [17]. Semaglutide clinical trials in Parkinson’s are underway. A 2025 JAMA Network Open study reported reduced neurodegeneration incidence with both semaglutide and tirzepatide in diabetes and obesity cohorts [18]. Our engine’s prediction — that these drugs work by recalibrating resource allocation rather than by targeting specific neuroprotective pathways — could explain why they show broad effects across multiple diseases simultaneously.

The ferroptosis connection is also worth noting. A 2024 review highlighted pollutant-induced ferroptosis as an emerging link between PD and cancer [4]. The engine’s “Redox State Lock-In” bridge predicted exactly this: the iron that powers neuronal respiration poisons the neuron when its metabolism fails, while cancer cells’ deliberate metabolic inefficiency shields them from the same poison.

5. Conclusion

We propose three contributions to the understanding of the Parkinson’s–cancer axis:

1. The Apoptotic Budget Hypothesis. The body maintains a finite, system-wide apoptotic capacity. Cancer and neurodegeneration are competing failures of this budget. This hypothesis generates testable predictions about apoptotic machinery throughput in disease states.

2. VDAC1 Competitive Docking Geometry. The cancer and neurodegeneration literatures on VDAC1 should be unified as a single competitive docking geometry problem. Drug design should target VDAC1 occupancy ratios, not VDAC1 function in isolation.

3. Resource Rationing as Drug Mechanism. Drugs like semaglutide, rapamycin, and metformin may work across the cancer–neurodegeneration boundary because they recalibrate the body’s resource allocation system, not because they target either disease specifically.

These findings were generated by a geometric knowledge discovery engine that does not read papers, does not follow established hypotheses, and does not know what the field considers plausible. It works from structure. And sometimes, structure sees what humans trained within disciplinary boundaries cannot.

All synthesis data, bridge tables, and engine output are available upon request. The engine’s geometric architecture is described in [7], prior work on alpha-synuclein is in [8], and geometric cancer topology is in [19].

References

[1] Bajaj, A., Driver, J.A. & Schernhammer, E.S. (2010). Parkinson’s disease and cancer risk: a systematic review and meta-analysis. Cancer Causes & Control, 21, 697–707.

[2] Zhang, X. et al. (2021). Parkinson’s disease and cancer: a systematic review and meta-analysis of over 17 million participants. BMJ Open, 11(9), e046329.

[3] El-Sayed, M.M. et al. (2025). Risk of colorectal cancer in Parkinson’s disease: a systematic review and meta-analysis of 11 million participants. BMC Neurology, 25, Article 206.

[4] Surguchev, A.A. & Surguchev, S. (2024). Association between Parkinson’s disease and cancer: new findings and possible mediators. International Journal of Molecular Sciences, 25(7), 3899.

[5] Inzelberg, R. & Flash, S. (2015). The associations between Parkinson’s disease and cancer: the plot thickens. Translational Neurodegeneration, 4, 20.

[6] Pan, T. et al. (2025). Parkinson’s disease and cancer: mechanistic insights and therapeutic opportunities from cancer neuroscience. MedComm – Oncology, e70044.

[7] Mekšriūnas, G. (2026). Subtractive Gap Detection in Geometric Knowledge Manifolds: Reverse Decomposition via Constraint Topology Conservation. Zenodo. Preprint, March 9, 2026.

[8] Mekšriūnas, G. (2026). Geometric Decomposition of the Alpha-Synuclein Aggregation Problem: Structural Prerequisites for Therapeutic Intervention in Parkinson’s Disease. Zenodo. Preprint, March 17, 2026.

[9] Magrì, A., Reina, S. & De Pinto, V. (2018). VDAC1 as pharmacological target in cancer and neurodegeneration: focus on its role in apoptosis. Frontiers in Chemistry, 6, 108.

[10] Magrì, A. et al. (2025). VDAC1: a key player in the mitochondrial landscape of neurodegeneration. Biomolecules, 15(1), 33.

[11] Magrì, A. et al. (2017). Interactions of VDAC with proteins involved in neurodegenerative aggregation: an opportunity for advancement on therapeutic molecules. Current Medicinal Chemistry, 24(40), 4470–4487.

[12] Shoshan-Barmatz, V. et al. (2015). At the crossroads between mitochondrial metabolite transport and apoptosis: VDAC1 as an emerging cancer drug target. Current Topics in Medicinal Chemistry.

[13] Wang, Y. et al. (2026). A small molecule VDAC ligand inhibits ERAD and induces selective cancer cell death via disruption of calcium homeostasis. Nature Communications, 17, 67816.

[14] Conti-Nibali, S. et al. (2026). Anti-cancer drugs targeting the NADH-binding site of VDAC rewire channel electrophysiology and partially suppress cation selectivity. The FEBS Journal, e70434.

[15] Gamborg, M. et al. (2025). GLP-1 agonists as potential neuromodulators in development of Parkinson’s disease: a nationwide cohort study. European Journal of Neurology, e70075.

[16] Zhang, L. et al. (2019). Semaglutide is neuroprotective and reduces alpha-synuclein levels in the chronic MPTP mouse model of Parkinson’s disease. Journal of Parkinson’s Disease, 9(1), 157–171.

[17] Meissner, W.G. et al. (2024). Lixisenatide in early Parkinson’s disease. New England Journal of Medicine, 390(13), 1176–1185.

[18] Nian, S. et al. (2025). Neurodegeneration and stroke after semaglutide and tirzepatide in patients with diabetes and obesity. JAMA Network Open, 8, e2521016.

[19] Mekšriūnas, G. (2026). The Topology of Malignancy: Metastasis as Holonomy, Dormancy as Spectral Trapping, and the Geometric Shape of Treatment Resistance. Zenodo. Preprint, March 11, 2026.

© 2026 Gedas Mekšriūnas / Omou Systems, MB. All rights reserved. omuo.io — Vilnius, Lithuania