Integrated Hypothesis: The Multi-Lock Trap
The hypotheses above are not mutually exclusive; indeed, the most compelling model for ME/CFS pathogenesis may involve multiple mechanisms operating simultaneously and reinforcing each other. We propose an integrated “multi-lock trap” hypothesis that attempts to explain the key features of ME/CFS: post-viral onset, persistence despite apparent resolution of the trigger, post-exertional malaise, multi-system involvement, and treatment resistance.
1 Phase 1: Triggering Event
An initial insult—typically viral infection, but potentially severe stress, trauma, or other immune-activating event—activates the evolutionarily conserved “sickness behavior” program. This is a normal, adaptive response involving:
- Metabolic downregulation (reduced mitochondrial activity, shifted fuel utilization)
- Immune activation and inflammatory cytokine production
- Behavioral changes (fatigue, social withdrawal, reduced activity)
- Tryptophan shunting toward kynurenine pathway
- Catecholamine conservation
In most individuals, this program disengages once the threat resolves. In ME/CFS-susceptible individuals, the program becomes “locked” through multiple overlapping mechanisms.
2 Phase 2: Lock Establishment
Several “locks” establish themselves during or shortly after the acute phase:
Epigenetic Lock. The severe metabolic stress creates stable epigenetic modifications in immune cells, neurons, muscle cells, and other tissues. Gene expression patterns appropriate for acute illness become fixed through DNA methylation and histone modifications. These changes persist through cell division, propagating the sick state even as acute inflammation resolves.
Autoimmune Lock. The inflammatory environment, possibly combined with molecular mimicry from the triggering pathogen, generates autoantibodies against self-proteins—G-protein coupled receptors, ion channels, or other cellular machinery. These autoantibodies create ongoing dysfunction independent of the original trigger. HERV reactivation during the acute phase may contribute immunogenic self-antigens.
Metabolic Lock. Tryptophan/kynurenine pathway dysregulation becomes self-perpetuating: inflammatory cytokines activate IDO, shunting tryptophan toward kynurenine; quinolinic acid accumulation causes neuroinflammation and oxidative stress; neuroinflammation maintains cytokine production, perpetuating IDO activation. Similar vicious cycles may establish in other metabolic pathways (lactate compartmentalization, purinergic signaling).
(Certainty: 0.50) We hypothesize that a self-reinforcing feedback loop linking kynurenine pathway activation, sleep disruption, mitochondrial dysfunction, and glutamate excitotoxicity contributes to ME/CFS pathogenesis and treatment resistance.
Proposed mechanism: Immune activation (e.g., post-viral) induces indoleamine 2,3-dioxygenase (IDO), shunting tryptophan toward kynurenine pathway production (Cervenka, Agudelo, and Ruas 2017),(Schwarcz et al. 2012). This reduces serotonin availability while increasing quinolinic acid (QA), a neurotoxic NMDA receptor agonist, and decreasing kynurenic acid (KYNA), a neuroprotective NMDA antagonist (Schwarcz et al. 2012). The QA:KYNA ratio elevation is observed in ME/CFS (Che et al. 2025),(Chojnacki et al. 2026).
Sleep disruption: Kynurenine accumulation disrupts sleep architecture, particularly slow-wave sleep (SWS) (Cervenka, Agudelo, and Ruas 2017). SWS is critical for glymphatic clearance—the brain’s waste removal system—which increases interstitial space by 60% and clears neurotoxic metabolites including β-amyloid, tau, and kynurenine metabolites (Xie et al. 2013). Unrefreshing sleep in ME/CFS may therefore cause metabolite accumulation, including QA and glutamate.
Mitochondrial dysfunction: The kynurenine shunt depletes NAD+ as tryptophan is diverted from serotonin to kynurenine-NAD+ synthesis, but this pathway is inefficient and produces reactive oxygen species (Cervenka, Agudelo, and Ruas 2017). NAD+ depletion impairs sirtuin activity, mitochondrial biogenesis, and oxidative phosphorylation (Che et al. 2025). ME/CFS patients show impaired energy production (citric acid cycle, β-oxidation, urea cycle) that worsens post-exercise (Che et al. 2025). Genetic evidence links mitochondrial genes AKAP10 and MTHFD1L to chronic fatigue (Z. Zhang et al. 2026).
Excitotoxicity via EAAT2 impairment: ATP depletion from mitochondrial dysfunction compromises Na+/K+ ATPase pumps, collapsing the sodium gradient essential for excitatory amino acid transporter 2 (EAAT2) function (Rothstein et al. 1996). EAAT2 is the primary glutamate reuptake transporter in astrocytes; its impairment causes extracellular glutamate accumulation and NMDA receptor overactivation (excitotoxicity) (Rothstein et al. 1996). In Alzheimer’s models, EAAT2 downregulation is directly linked to glutamate accumulation and neuronal damage (El-Baga et al. 2026),(Manisha et al. 2026),(X. Zhang et al. 2026).
Feedback amplification: QA directly activates NMDA receptors, causing calcium influx, oxidative stress, and further mitochondrial damage (Schwarcz et al. 2012). Excitotoxicity amplifies neuroinflammation, which maintains IDO activation and perpetuates kynurenine production. Glymphatic impairment from sleep disruption reduces clearance of QA, glutamate, and other neurotoxic metabolites, creating a self-sustaining cycle.
Evidence base: Multi-omics studies show kynurenine pathway dysregulation in ME/CFS, with abnormalities worsening post-exercise (Che et al. 2025). A clinical trial found elevated QA and reduced KYNA in CFS+IBS patients, with probiotic treatment improving the QA:KYNA ratio by 45% and reducing fatigue by 40.3% (Chojnacki et al. 2026). Sleep architecture disruption is a core ME/CFS symptom; glymphatic impairment provides a mechanistic link to metabolite accumulation (Xie et al. 2013). Mitochondrial dysfunction is well-documented in ME/CFS muscle and blood (Fanò-Illic et al. 2026),(Z. Zhang et al. 2026). EAAT2 impairment and excitotoxicity are established in neurodegenerative diseases; extrapolation to ME/CFS is mechanistically plausible but untested (El-Baga et al. 2026),(Manisha et al. 2026),(X. Zhang et al. 2026).
Testable predictions:
- ME/CFS patients show elevated QA:KYNA ratio compared to healthy controls, with higher ratio correlating with symptom severity (Chojnacki et al. 2026)
- Sleep polysomnography reveals reduced SWS percentage; SWS deficit correlates with CSF QA and glutamate levels
- ATP depletion in peripheral blood mononuclear cells correlates with EAAT2 expression (if measurable in astrocyte-derived exosomes)
- Combination treatment targeting kynurenine pathway (e.g., probiotics), sleep quality, and mitochondrial function shows greater efficacy than monotherapy
- Post-exercise exacerbation coincides with transient worsening of QA:KYNA ratio and glutamate clearance
Therapeutic implications: This hypothesis suggests multi-target intervention: (1) Kynurenine pathway modulation—probiotics to shift gut-kynurenine axis toward KYNA (Chojnacki et al. 2026), IDO inhibitors, or KAT activators; (2) Sleep enhancement—pharmacological or behavioral interventions to increase SWS; (3) Mitochondrial support—NAD+ precursors, CoQ10, antioxidants; (4) EAAT2 enhancement—agents that upregulate EAAT2 expression or function (extrapolated from Alzheimer’s research) (Manisha et al. 2026).
Limitations: No direct EAAT2 studies in ME/CFS; all EAAT2 evidence extrapolated from Alzheimer’s models (El-Baga et al. 2026),(Manisha et al. 2026),(X. Zhang et al. 2026). Glymphatic function has not been directly measured in ME/CFS; sleep architecture studies are limited. The feedback loop is mechanistically plausible but not yet demonstrated as a unified entity in ME/CFS. Some evidence is contradictory—e.g., animal fatigue models show kynurenine pathway inhibition rather than activation (Yan et al. 2026), suggesting pathway dysregulation may be bidirectional or context-dependent.
Replication status: Kynurenine pathway dysregulation in ME/CFS partially replicated (Che2025, Chojnacki2026 show consistent QA:KYNA findings). Sleep-glymphatic link fully replicated across multiple diseases (Xie2013 and subsequent studies). Mitochondrial dysfunction in ME/CFS extensively replicated. EAAT2 findings not replicated in ME/CFS (extrapolated from Alzheimer’s).
(Certainty: 0.55) Individual variation in glymphatic efficiency may determine ME/CFS trajectory. The kynurenine-excitotoxicity loop predicts toxic metabolite accumulation (QA, glutamate). Glymphatic clearance during slow-wave sleep (SWS) is the primary mechanism for removing these metabolites (Xie et al. 2013). Patients with lower glymphatic clearance rates (measured by DTI-ALPS) may fail to clear neurotoxic metabolites to tolerable levels, allowing loop stabilization and chronic pathology. Patients with higher clearance rates may clear metabolites sufficiently to allow recovery, even with similar initial insult.
Mechanistic rationale: ME/CFS patients show variable recovery trajectories from viral triggers; some resolve spontaneously while others develop chronic ME/CFS (Walitt et al. 2024). If toxic metabolite accumulation is the loop’s amplifying factor, glymphatic efficiency could be the critical differentiator. Sleep disruption (core ME/CFS symptom) directly reduces glymphatic clearance (Xie et al. 2013). DTI-ALPS (diffusion tensor imaging along perivascular space) is a validated, non-invasive measure of glymphatic function in other neurological disorders (Xie et al. 2013).
Testable predictions:
- ME/CFS patients show lower DTI-ALPS index (higher glymphatic clearance) than healthy controls
- Within ME/CFS cohort, lower ALPS index predicts: (a) higher CSF QA concentration, (b) reduced SWS percentage on polysomnography, (c) slower recovery from symptom exacerbation after standardized cognitive challenge
- DTI-ALPS improves after interventions known to enhance glymphatic function (sleep enhancement, positional therapy)
Limitations: Glymphatic function not directly measured in ME/CFS; DTI-ALPS validated in other conditions but not yet applied to ME/CFS. Sleep architecture studies in ME/CFS are limited. This hypothesis correlates two untested variables (DTI-ALPS and CSF QA) in ME/CFS patients.
Replication status: Sleep-glymphatic link fully replicated across multiple diseases (Xie2013 and subsequent studies). DTI-ALPS measurement replicated as glymphatic function marker in Alzheimer’s, normal pressure hydrocephalus, and other conditions. ME/CFS-specific validation lacking.
(Certainty: 0.45) The kynurenine-excitotoxicity loop has multiple amplification points, but the ATP-dependent EAAT2 glutamate transporter may be the most fragile: it requires both adequate ATP (from mitochondrial function) and intact Na+ gradient (from Na+/K+ pump). This dual dependence creates a “bottleneck” where small mitochondrial deficits cascade into large excitotoxicity increases (Rothstein et al. 1996).
Mechanistic rationale: EAAT2 requires Na+ gradient maintained by Na+/K+ ATPase pumps; these pumps are ATP-dependent. Mitochondrial dysfunction in ME/CFS reduces ATP production (Che et al. 2025),(Fanò-Illic et al. 2026). In ME/CFS, mitochondrial genes AKAP10 and MTHFD1L are causally linked to chronic fatigue (Z. Zhang et al. 2026). If ATP depletion is the primary driver of EAAT2 impairment, even partial mitochondrial failure could cause disproportionate glutamate accumulation. In Alzheimer’s models, EAAT2 downregulation directly causes glutamate accumulation and neuronal damage (El-Baga et al. 2026),(Manisha et al. 2026). ATP depletion may not fully collapse EAAT2 (Na+ gradient may persist), but reduced function could still cause clinically relevant excitotoxicity.
Testable predictions:
- In ME/CFS patients, EAAT2 function (measured in astrocyte-derived exosomes or CSF) shows stronger correlation with mitochondrial ATP production (measured in PBMCs) than with neuroinflammatory markers
- Pharmacological Na+/K+ pump stabilizers (sub-toxic digitalis-like compounds at doses insufficient for CNS effects) improve glutamate clearance independent of mitochondrial function
- EAAT2-targeted interventions show greater efficacy than mitochondrial support alone, suggesting EAAT2 is the limiting factor
Limitations: No direct EAAT2 studies in ME/CFS; all EAAT2 evidence extrapolated from Alzheimer’s models (El-Baga et al. 2026),(Manisha et al. 2026). Exosome-based EAAT2 measurement in ME/CFS not yet developed. The relationship between ATP levels and EAAT2 function in ME/CFS is untested.
Replication status: EAAT2-ATP coupling demonstrated in Rothstein1996 (knockout mice). Mitochondrial dysfunction in ME/CFS extensively replicated. EAAT2 downregulation in neurodegenerative diseases replicated. ME/CFS-specific coupling evidence lacking.
(Certainty: 0.40) Contradictory findings in literature may reflect time-dependent biphasic response: acute inflammation triggers IDO activation and kynurenine surge; chronic activation depletes tryptophan pool and impairs downstream enzymes (KAT, KMO), causing metabolite imbalance. ME/CFS patients are in depleted phase but with elevated QA:KYNA ratio due to differential enzyme sensitivity (Che et al. 2025),(Yan et al. 2026).
Mechanistic rationale: ME/CFS studies show kynurenine pathway activation with elevated QA and reduced KYNA (Che et al. 2025),(Chojnacki et al. 2026). Animal chronic fatigue models show kynurenine pathway inhibition rather than activation, with Dendrobium officinale treatment restoring pathway activity (Yan et al. 2026). This contradiction may reflect biphasic disease progression: acute post-viral phase shows IDO upregulation and kynurenine surge; chronic depletes tryptophan (substrate limitation) and impairs KAT/KMO enzymes, causing persistent QA:KYNA imbalance. Inflammation-driven IDO activation persists, but metabolic capacity to convert kynurenine to KYNA declines, trapping metabolism in neurotoxic QA-dominant state.
Testable predictions:
- In longitudinal study of acute viral infection followed to chronic phase: kynurenine:tryptophan ratio peaks at 1–3 months post-infection in those who develop ME/CFS
- Ratio declines by 6–12 months but QA:KYNA ratio remains elevated in ME/CFS patients
- Enzyme expression shows temporal pattern: IDO upregulated early, KAT downregulated late in ME/CFS patients vs. resolved infection controls
- Treatment timing matters: IDO inhibitors beneficial early (acute phase), KAT activators beneficial late (chronic phase)
Limitations: Biphasic hypothesis untested in ME/CFS; longitudinal data from acute viral trigger through chronic ME/CFS not available. Human data on KAT/KMO enzyme changes over time in ME/CFS lacking. Animal model (Yan2026) shows inhibition in chronic fatigue, but Dendrobium is traditional medicine with broader metabolic effects; may not directly translate to human ME/CFS.
Replication status: Kynurenine pathway activation in ME/CFS partially replicated (Che2025, Chojnacki2026). Kynurenine pathway inhibition in chronic fatigue animal model (Yan2026) is single study. Longitudinal data lacking in ME/CFS.
Signaling Lock. Purinergic receptors become sensitized, vagal afferents develop persistent danger signaling, or cellular quorum sensing becomes corrupted. The body’s communication systems now interpret normal physiological states as pathological.
Structural Lock. Glymphatic impairment, circadian desynchronization, or redox compartment collapse creates physical or temporal barriers to normal function that resist simple correction.
3 Phase 3: Trap Maintenance
Once multiple locks are established, the system becomes trapped in a stable pathological state. Each lock reinforces the others:
- Epigenetic changes maintain cells in a “sickness program” gene expression state
- Autoantibodies cause ongoing receptor/channel dysfunction
- Metabolic pathway dysregulation depletes essential intermediates while accumulating toxic ones
- Aberrant signaling maintains central nervous system perception of threat
- Structural/temporal disruptions prevent normal clearing and cycling
Attempting to force the system out of this state (through exertion, stimulants, or willpower) triggers defensive responses: the body “detects” that something is trying to override its protective program during perceived danger, and responds by intensifying the sickness response—post-exertional malaise.
4 Why Recovery Is Rare
For recovery to occur, all locks must be released, or at least enough of them that the remaining ones cannot maintain the trapped state. Treatments targeting only one mechanism fail because the others maintain the trapped state. This explains why:
- Immunomodulation sometimes helps but rarely cures (addresses autoimmune lock only)
- Metabolic supplements show limited efficacy (addresses metabolic lock only)
- Behavioral approaches fail or cause harm (don’t address any locks, may strengthen them)
- Early intervention shows better outcomes (fewer locks have stabilized)
- Spontaneous recovery is rare and unpredictable (requires spontaneous release of multiple locks)
- Some patients respond to treatments others don’t (different lock combinations)
5 Testable Predictions
This integrated hypothesis generates several testable predictions:
- ME/CFS patients should show epigenetic signatures distinct from healthy controls and from recovered patients, potentially with duration-dependent stabilization
- Multiple autoantibody classes should be present, not just one type
- Kynurenine pathway metabolites should show specific patterns (elevated quinolinic:kynurenic ratio)
- Purinergic receptor expression or sensitivity should differ from controls
- Combined treatments targeting multiple locks should show synergistic efficacy compared to monotherapies
- Patients who recover should show reversal of epigenetic changes, autoantibody clearance, or both
- Disease duration should correlate with epigenetic change stability and treatment resistance
- Patient subgroups might be identifiable by which locks predominate
- Kynurenine-excitotoxicity loop-specific predictions:
- ME/CFS patients show elevated QA:KYNA ratio compared to healthy controls; higher ratio correlates with symptom severity (Chojnacki et al. 2026)
- Sleep polysomnography reveals reduced SWS percentage; SWS deficit correlates with CSF QA and glutamate levels (Xie et al. 2013)
- ATP depletion in peripheral blood mononuclear cells correlates with EAAT2 expression (if measurable in astrocyte-derived exosomes) (Rothstein et al. 1996),(El-Baga et al. 2026),(Manisha et al. 2026),(X. Zhang et al. 2026)
- Combination treatment targeting kynurenine pathway, sleep quality, and mitochondrial function shows greater efficacy than monotherapy (Cervenka, Agudelo, and Ruas 2017),(Xie et al. 2013),(Che et al. 2025)
- Post-exercise exacerbation coincides with transient worsening of QA:KYNA ratio and glutamate clearance (Che et al. 2025)
- Glymphatic clearance rate (DTI-ALPS index) inversely correlates with CSF QA concentration and fatigue severity (Xie et al. 2013),(Che et al. 2025)
- Mitochondrial-EAAT2 coupling: EAAT2 function shows stronger correlation with mitochondrial ATP production than with neuroinflammatory markers (Rothstein et al. 1996),(El-Baga et al. 2026),(Manisha et al. 2026),(Z. Zhang et al. 2026)
- Biphasic kynurenine dysregulation: Longitudinal studies show IDO upregulation early (1–3 months post-infection) with KAT downregulation later (6–12 months) in ME/CFS patients (Che et al. 2025),(Yan et al. 2026)
- Baseline microbiome composition predicts probiotic response: patients with low Bifidobacterium and high Proteobacteria show greater improvement in QA:KYNA ratio after CDS22 probiotic (Chojnacki et al. 2026)
- NAD+ sirtuin threshold: Bimodal NAD+ distribution in ME/CFS PBMCs (normal cluster vs critically low cluster); low-NAD+ cluster shows reduced sirtuin activity and correlates with disease duration (Cervenka, Agudelo, and Ruas 2017),(Che et al. 2025)
The multi-lock trap model is a conceptual framework, not a falsifiable hypothesis in its current form. Key epistemic boundaries:
- The model can accommodate virtually any finding by positing an additional “lock”—this explanatory flexibility is a weakness, not a strength, as it makes the model difficult to disprove.
- No study has simultaneously measured epigenetic, autoimmune, metabolic, signalling, and structural “locks” in the same ME/CFS cohort; the integration is theoretical, not data-driven.
- The prediction that “combined treatments targeting multiple locks should show synergistic efficacy” has not been tested; it is equally possible that the locks are independent phenomena requiring independent treatment, or that addressing a single upstream driver resolves multiple downstream manifestations.
- The model’s value lies in organising research priorities and treatment strategies, not in establishing causal architecture.
6 Therapeutic Implications
If the multi-lock model is correct, effective treatment would require simultaneously addressing multiple mechanisms:
- Epigenetic modifiers: Agents that can reverse pathological epigenetic programming (HDAC inhibitors, DNA demethylating agents, or lifestyle interventions that affect the epigenome)
- Autoantibody reduction: Plasmapheresis, rituximab, IVIG, or tolerization approaches
- Metabolic pathway correction: Targeted supplementation to restore normal flux through kynurenine and other pathways; NAD+ precursors; specific nutrient support
- Kynurenine pathway modulation (Tier 1):
- Probiotic: CDS22 multi-strain probiotic (450B CFU/day, 12 weeks) to shift gut-kynurenine axis toward KYNA, improve QA:KYNA ratio by 45% and reduce fatigue by 40.3% (Chojnacki et al. 2026)
- NAD+ precursors (Tier 1): Nicotinamide riboside (NR) or nicotinamide mononucleotide (NMN) 250–500mg daily to bypass kynurenine shunt, restore mitochondrial function, and reactivate sirtuins (Cervenka, Agudelo, and Ruas 2017),(Che et al. 2025)
- NMDA antagonists (Tier 2): Dextromethorphan or memantine to block QA-NMDA receptor activation, reducing excitotoxicity upstream (Schwarcz et al. 2012),(El-Baga et al. 2026)
- KAT activators (Tier 1): Agents that increase kynurenine aminotransferase activity to shift metabolism toward KYNA (neuroprotective) rather than QA (neurotoxic) (Schwarcz et al. 2012)
- Na+/K+ pump stabilizers (Tier 2): Low-dose digitalis-like compounds to preserve ATPase pump function independent of mitochondrial output, supporting EAAT2-mediated glutamate clearance (Rothstein et al. 1996)
- Sleep enhancement (Tier 1):
- SWS enhancement protocols: Pharmacological (e.g., trazodone at low dose) or behavioral interventions (sleep hygiene, temperature optimization) to increase slow-wave sleep percentage, enhancing glymphatic clearance (Xie et al. 2013)
- Positional sleep therapy: Sleeping with head elevated 10–15° to leverage gravity-driven interstitial fluid exchange, improving glymphatic clearance independent of sleep architecture
- Glymphatic sleep bundle: Combine SWS enhancement with positional therapy and circadian alignment to maximize metabolite clearance during sleep (Xie et al. 2013)
- Mitochondrial support (Tier 1):
- CoQ10 100–300mg daily, alpha-lipoic acid 300–600mg daily, and acetyl-L-carnitine 500–2000mg daily to support oxidative phosphorylation
- Mitochondrial biogenesis support: PQQ 20mg daily, ALCAR 500–1000mg daily to stimulate new mitochondria formation
- EAAT2 enhancement (Tier 2):
- PirB-mTOR pathway modulators: Agents that inhibit PirB to increase EAAT1/EAAT2 expression via mTOR activation, enhancing glutamate clearance (X. Zhang et al. 2026)
- Redox support: N-acetylcysteine (NAC) 600–1800mg daily to maintain EAAT2 redox state and prevent oxidative impairment (Manisha et al. 2026),(El-Baga et al. 2026)
- Lion’s mane (Hericium erinaceus) 500–3000mg daily to upregulate EAAT2 expression in astrocytes (Manisha et al. 2026)
- Signaling normalization: Purinergic receptor antagonists, vagal nerve modulation, low-dose naltrexone (affects multiple signaling systems)
- Structural/temporal restoration: Addressing craniocervical issues, chronotherapy for circadian resynchronization, targeted redox support
- Pacing and energy management: Preventing exertion-triggered lock reinforcement while other interventions work
The timing and sequencing of interventions may matter: some locks may need to be addressed before others become accessible. For example, reducing autoantibodies might be necessary before epigenetic interventions can take effect.
Combination protocols (Tier 1/2):
- Glymphatic sleep bundle: SWS enhancement (pharmacological or behavioral) + positional sleep therapy (elevation 10–15°) + circadian alignment (morning light exposure, consistent schedule) to maximize glymphatic clearance during sleep (Xie et al. 2013)
- Kynurenine stack: Probiotic (CDS22 450B CFU/day) + NAD+ precursors (NR 250mg daily) + KAT cofactors (vitamin B6 50mg, P5P 100mg daily) to shift kynurenine metabolism toward KYNA and restore energy metabolism (Chojnacki et al. 2026),(Cervenka, Agudelo, and Ruas 2017),(Schwarcz et al. 2012)
- Mitochondrial-EAAT2 protocol: CoQ10 200mg + alpha-lipoic acid 400mg + NAC 1200mg + Na+/K+ pump stabilizer (sub-toxic digitalis-like compound) to support ATP production, reduce EAAT2 oxidative impairment, and preserve pump function independent of mitochondrial output (Rothstein et al. 1996),(El-Baga et al. 2026),(Manisha et al. 2026)
- Vagal-sleep-mitochondrial triad: tVNS 15min BID (gammaCore or Nemos device) for parasympathetic activation and cholinergic anti-inflammatory pathway + SWS enhancement protocol + mitochondrial support (NR 250mg + CoQ10 200mg) to address three systems simultaneously (Cervenka, Agudelo, and Ruas 2017),(Xie et al. 2013),(Schwarcz et al. 2012)
- Stratified protocol (Tier 1): Use baseline biomarker profile (QA:KYNA ratio, microbiome composition, ATP:NAD+ ratio) to guide treatment selection: probiotic-responsive vs. metabolic-focused vs. neuroinflammatory-focused (Chojnacki et al. 2026),(Cervenka, Agudelo, and Ruas 2017),(Che et al. 2025)
- Rescue protocol for flares (Tier 1): Acute PEM exacerbation protocol: (a) reduce exertion to 50% of baseline for 48–72 hours, (b) NAC 600mg + magnesium 400mg + vitamin B6 50mg for EAAT2 redox support, (c) sleep optimization (dark room, elevated sleep position), (d) avoid immune triggers (infections, stress) during flare (Rothstein et al. 1996),(Cervenka, Agudelo, and Ruas 2017)
7 Research Directions
This model suggests several research priorities:
- Comprehensive phenotyping: Assessing each patient for multiple lock types to enable personalized treatment
- Combination therapy trials: Testing whether multi-target approaches show synergy
- Longitudinal tracking: Following lock status over time to understand disease progression and treatment effects
- Early intervention studies: Testing whether aggressive early treatment can prevent lock stabilization
- Recovery studies: Detailed analysis of the rare patients who recover to understand which locks released and how
- Kynurenine pathway-specific studies:
- Glymphatic clearance assessment: DTI-ALPS (diffusion tensor imaging along perivascular space) to quantify glymphatic efficiency; correlate with CSF QA and glutamate levels and SWS percentage on polysomnography (Xie et al. 2013)
- Mitochondrial-EAAT2 coupling: Develop assay measuring EAAT2 function in astrocyte-derived exosomes; correlate with mitochondrial ATP production in PBMCs to test whether ATP depletion is primary EAAT2 driver (Rothstein et al. 1996),(El-Baga et al. 2026),(Manisha et al. 2026)
- Probiotic stratification: Replicate Chojnacki2026 with RCT design; stratify patients by baseline microbiome (Bifidobacterium vs Proteobacteria) to predict probiotic response and QA:KYNA ratio improvement (Chojnacki et al. 2026)
- Longitudinal kynurenine profiling: Track kynurenine pathway enzymes (IDO, KAT, KMO) and metabolites (QA, KYNA, 3-HK) from acute phase through chronic ME/CFS to test biphasic dysregulation hypothesis (Che et al. 2025),(Yan et al. 2026)
- Muscle clearance capacity: Measure kynurenine clearance in skeletal muscle (enzyme expression, mitochondrial function) vs. systemic kynurenine levels to test exercise intolerance mechanism (Cervenka, Agudelo, and Ruas 2017),(Fanò-Illic et al. 2026)
- Sirtuin threshold assessment: Measure NAD+ levels and sirtuin activity in PBMCs; test for bimodal distribution correlating with disease duration and treatment response (Cervenka, Agudelo, and Ruas 2017),(Che et al. 2025)
- Brainstem glutamate assessment: MR spectroscopy targeting NTS and vagal nuclei to measure glutamate concentrations; correlate with dysautonomic symptoms (POTS, orthostatic intolerance) (Schwarcz et al. 2012)
- KMO redox sensitivity: Test whether KMO enzyme activity is impaired by oxidative stress in ME/CFS; assess whether antioxidants restore KMO function and shift metabolism away from QA (Schwarcz et al. 2012)
- Glymphatic clearance correlation: Direct measurement of glymphatic efficiency (DTI-ALPS) alongside CSF metabolite profiling (QA, glutamate) and sleep architecture (SWS percentage) to establish mechanistic links between sleep quality and neurotoxic metabolite accumulation (Xie et al. 2013),(Che et al. 2025)
- EAAT2-based therapeutic screening: Test Na+/K+ pump stabilizers (sub-toxic digitalis-like compounds) for ability to enhance EAAT2 function independent of ATP; compare efficacy to mitochondrial support alone (Rothstein et al. 1996),(El-Baga et al. 2026),(Manisha et al. 2026),(X. Zhang et al. 2026)