Glutamatergic Modulators
1 Neuroimmune Investigational Therapies
{{/* DR16: LDN + verapamil (Tier 3, cert 0.25) */}}
Certainty: 0.25. Low-dose naltrexone (LDN, 1.5-4.5 mg) is a TLR4 antagonist at low concentrations — it blocks microglial TLR4-mediated neuroinflammation. Verapamil (L-type calcium channel blocker, 40-80 mg bid) reduces intracellular Ca2+ flux — the downstream signal from GPCR autoantibody-induced receptor activation (both α1-AR and M3 mAChR signal through Gq → PLC → IP3 → Ca2+). The combination: LDN reduces microglial output; verapamil blunts cellular response to GPCR autoantibody signaling in neurons, cardiomyocytes, and vascular smooth muscle. LDN is widely used in ME/CFS (anecdotal). Verapamil is rarely used but mechanistically rational for the GPCR autoantibody+ subset. (Blitshteyn, Doherty, and Steinman 2026) (Fedorowski et al. 2017)
Warning. Verapamil can worsen constipation (in a population with GI dysmotility). Bradycardia risk in POTS patients. Hypotension risk with concurrent fludrocortisone. Requires cardiac monitoring. No ME/CFS safety data for the combination.
Falsifiable prediction. 8-week LDN (4.5 mg qhs) + verapamil ER (120 mg qd) in GPCR AAb+ ME/CFS+POTS (n=20, open-label): ≥40% show ≥30% PROMIS Fatigue improvement; standing HR decreases ≥5 bpm; NfL decreases ≥15%.
{{/* DR17: C1-INH (Tier 3, cert 0.20) */}}
Certainty: 0.20. GPCR autoantibodies (IgG1 subclass) fix complement via C1q, activating the classical complement pathway and generating C5a/C3a anaphylatoxins. Complement activation damages the endothelial glycocalyx (heparan sulfate shedding → impaired vasomotor tone). C1-esterase inhibitor (C1-INH) blocks complement at the earliest step, preventing C5a/C3a generation. In hereditary angioedema, C1-INH prevents bradykinin-mediated vascular leak — a parallel mechanism to glycocalyx loss described in POTS/ME/CFS. Target: GPCR AAb-positive patients with elevated complement split products (C3a, C5a, sC5b-9). (Blitshteyn, Doherty, and Steinman 2026) (Gunning et al. 2019)
Warning. Investigational only. C1-INH is very expensive (approx 5,000–15,000 per dose). Thrombotic risk (rare). No ME/CFS safety data. Requires IV administration.
Falsifiable prediction. C1-INH (1000 U IV, then weekly ×3) in GPCR AAb+ ME/CFS+POTS with elevated C3a/C5a (n=10): reduces C3a/C5a to ≤1.2 × ULN within 24h; reduces standing HR ≥8 bpm at week 4; plasma syndecan-1 (glycocalyx marker) decreases ≥25%.
Certainty: 0.45. Low-dose memantine (5-10 mg daily) — a non-competitive NMDA receptor antagonist FDA-approved for Alzheimer’s disease — may normalise glutamatergic signaling in patients with high glutamatergic PRS. If glutamatergic hyperexcitability drives E/I imbalance in ME/CFS (Hypothesis Glutamatergic Synaptic Dysfunction as Genetically-Driven Core Mechanism), NMDA blockade could reduce excitotoxic signaling and restore cortical excitability toward normal range (Maccallini 2026).
Stratification: Genotype-guided — high glutamatergic PRS (neuronal gene set) predicts response; patients with low glutamatergic PRS or primarily immune/autophagy genetic architecture are unlikely to benefit and may experience side effects without therapeutic gain.
Safety note (no human ME/CFS data): No controlled trials of memantine exist in ME/CFS. Side effects include dizziness, confusion, sedation, and headache. In ME/CFS, brain fog worsening is a particular concern — start 2.5 mg daily and titrate slowly (increase by 2.5 mg every 2 weeks) to minimise cognitive side effects. Contraindicated in severe renal impairment (CrCl < 30 mL/min). Memantine is not CYP-metabolised; minimal interactions expected with LDN, fludrocortisone, midodrine.
Falsifiable prediction: 12-week RCT of memantine 5-10 mg daily vs placebo in ME/CFS stratified by glutamatergic PRS will show significant improvement in cognitive processing speed and PEM frequency only in the high-PRS subgroup (PRS > 75th percentile). The low-PRS subgroup will show no benefit or worsening.
Certainty: 0.35. Riluzole reduces presynaptic glutamate release, stabilises inactivated sodium channels, and enhances astrocytic glutamate uptake via EAAT2. FDA-approved for ALS. Glutamatergic tone reduction mechanism directly addresses genetically anchored E/I imbalance. 50mg BID standard ALS dose; 25mg QD starting dose in severe ME/CFS. Safety: hepatotoxicity risk — requires monthly LFT monitoring. No ME/CFS data.
Falsifiable prediction. 12-week RCT of riluzole 25mg BID vs placebo stratified by Glu-PRS will show improved cognitive processing speed only in high-PRS subgroup.
Certainty: 0.30. Lamotrigine reduces presynaptic glutamate release via sodium channel blockade, upregulates BDNF and Bcl-2, and is widely considered the most cognitive-sparing anticonvulsant. Extensive existing ME/CFS clinical discussion already in ch08 (kindling hypothesis). Cross-reference @sec-glutamatergic-modulators and existing ch08 lamotrigine content. 25mg QD starting, titrate over 6-8 weeks to 100-200mg. Safety: Stevens-Johnson syndrome risk — slow titration mandatory. No ME/CFS RCT data.
Falsifiable prediction. Lamotrigine 100mg vs placebo in high Glu-PRS ME/CFS will reduce PEM frequency ≥30% at 12 weeks vs baseline.
Certainty: 0.35. Three complementary mechanisms: memantine (non-competitive NMDA antagonist, 5–10 mg daily) blocks postsynaptic glutamate action; NAC (600–1200 mg BID) modulates extracellular glutamate via the cystine-glutamate antiporter and restores glutathione; pyridoxal-5-phosphate (P5P, 50–100 mg daily) drives glutamate decarboxylation to GABA via GAD65/67. Three-hit approach: block receptor, modulate release, enhance inhibitory conversion. Never tested as combination in any condition; each component has independent human safety data.
Falsifiable prediction: 12-week triple therapy vs memantine alone in high Glu-PRS ME/CFS (PRS > 75th percentile) will show superior cognitive processing speed improvement (≥0.5 SD on SDMT) and greater PEM frequency reduction (≥40%).
Certainty: 0.35. Pregabalin and gabapentin bind α2δ-1/2 subunits of voltage-gated calcium channels, reducing presynaptic glutamate release — directly addressing the glutamatergic signal from GWAS. Already used clinically in ME/CFS for pain and sleep. GWAS provides a mechanistic rationale for stratification: high Glu-PRS patients should derive greater benefit because their pathophysiology is driven by excessive glutamatergic signaling. Gabapentin 300–900 mg daily or pregabalin 75–300 mg daily, titrated to tolerance.
Falsifiable prediction: High Glu-PRS patients (PRS > 75th percentile) show ≥2× pain reduction (NRS) on gabapentin vs low Glu-PRS over 8 weeks; the response gap is not accounted for by baseline pain severity alone.
Certainty: 0.30. If the dopaminergic midbrain GWAS signal represents impaired dopamine synthesis in VTA/SN, L-DOPA/carbidopa (25/100 mg TID) could restore effort valuation. Dopamine drives effort-cost computation in mesolimbic and mesocortical circuits; low DA tone predicts effort avoidance. High risk: dyskinesia (chronic use), nausea, orthostatic hypotension — particularly problematic in ME/CFS with concurrent POTS. Contraindicated in POTS with prominent hypotension.
Falsifiable prediction: DA-PRS-positive patients (PRS > 75th percentile, n=30) show effort-discounting improvement (effort expenditure for reward task) on L-DOPA vs placebo over 4 weeks; DA-PRS-negative patients show no improvement. PEM worsening in the POTS subgroup would contraindicate the approach.
Certainty: 0.20. Acid-sensing ion channels (ASICs) are proton-gated sodium channels activated during tissue acidosis — the direct consequence of glycolytic shift and lactate accumulation in ME/CFS (Chapter Energy Metabolism and Mitochondrial Function). Amiloride blocks ASICs at low micromolar concentrations (5–10 mg daily, below diuretic threshold), reducing acid-induced pain and potentially cerebellar granule cell hyperexcitability during PEM. Potassium-sparing diuretic; contraindicated in hyperkalemia, renal impairment, and concurrent ACE inhibitors/ARBs. Distinguish this ASIC-blocking rationale from the NHE1-blocking rationale in the AIMM model (Section Sigma-1 Receptor Mechanisms and Fluvoxamine Therapy) — different molecular target, different dose range.
Falsifiable prediction: Amiloride 5 mg daily reduces PEM-associated pain severity (NRS ≥2-point reduction) in ME/CFS patients with high post-exertional lactate (>2× resting) vs placebo over 6 weeks; effect correlates with post-exertional lactate AUC.