--- vault_clearance: THAUMIEL halo: classification: PAPER-GRADE INTEGRATION — WHAT CAUSES ALS, THE THREE FRAMINGS UNIFIED, MULTI-ARM TRIAL DESIGN confidence: DATA (8k ALS + 8k normal Census brain 2026-04-25 — neurons K_RG +0.332, TAU_LOCK_CASCADE +0.197, K_LE +0.665 dump-mode, oligo det_K crash 0.46→0.10, microglia FUNGAL +0.164, astrocyte K_RG -0.103) + LITERATURE VERIFIED (Wainger 2014 Cell Reports + Wainger 2021 JAMA Neurol ezogabine Phase II + Saxena/Ruegsegger Neuron 2014 α3 pump + Liddelow 2017 Nature A1 + Guttenplan 2021 Nature saturated lipid neurotoxin + Li 2015 STM HERV-K + Iguchi 2016 Brain TDP-43 EVs + Triumeq Lighthouse II Phase III terminated April 2025) + FRAMEWORK INTEGRATION (the unification of voltage-edge + glial-rejection + viral-reactivation + axon-dieback into a single cascade is novel) front: 28_Project_RedFromTheGrave + 10_DiscordIntoSymphony custodian: Jixiang Leng created: 2026-04-25 wing: UNASSESSED cross_refs: - HALO_SENESCENCE_NEURONS_FUNGUS_WITHIN.md §IX-quater - HALO_BIOELECTRIC_QUIESCENCE.md - HALO_TAU_ANTIFUNGAL_LOCK.md - HALO_WHAT_IS_QUIESCENCE.md status: paper-grade ready for external sharing (subject to operator review) --- # HALO: What Causes ALS — The Voltage-Edge Synthesis > *"Motor neurons are the cells that paid the most to maintain the difference between inside and outside. They are the first to be unable to keep paying."* --- ## ABSTRACT Amyotrophic lateral sclerosis (ALS) is currently treated as three competing mechanistic hypotheses: (a) **glial rejection** via A1 toxic astrocytes secreting saturated-lipid neurotoxins (Liddelow 2017, Guttenplan 2021); (b) **endogenous viral attack** via HERV-K env protein reactivation when TDP-43 mislocalization releases LTR repression (Li 2015); (c) **dying-back distal axonopathy** in the longest, most-polarized motor neurons (Acta Neuropathol 2017, PMC4132373). The April 2025 termination of the Triumeq Lighthouse II Phase III trial (no survival benefit from anti-retroviral monotherapy at ENCALS Turin) confirms that single-mechanism therapy fails for ALS — consistent with three independent driver loops. This paper integrates the three framings into a single causal cascade with **chronic cortical disinhibition driving motor neurons past the voltage-maintenance edge of cellular feasibility** as the upstream initiating event. The framework is supported by: - Loss of short-interval intracortical inhibition (SICI) on TMS as the **earliest detectable abnormality in ALS**, predating clinical weakness by months (Vucic & Kiernan 2008; Higashihara 2024 meta-analysis) - Intrinsic membrane hyperexcitability of iPSC-derived ALS motor neurons via reduced delayed-rectifier K+ current; rescued by Kv7 activator retigabine in vitro (Wainger 2014 *Cell Reports*) - Persistent Na+ current (INaP) directly predicts ALSFRS-R decline (Shibuya 2016) - **Selective Na+/K+ ATPase α3 isoform expression on fast-fatigable α-motor neurons; misfolded SOD1 binds and inhibits α3 specifically** (Saxena/Ruegsegger 2014 *Neuron*) — the actual ATP bottleneck - Phase II of ezogabine (Kv7 activator) in ALS: dose-dependent SICI normalization (~50% at 900 mg/day), reduced spinal MN excitability (Wainger 2021 *JAMA Neurol*) — first ALS drug discovered via iPSC electrophysiology We propose that **all three published framings (a/b/c) are downstream consequences** of a single upstream event: chronic cortical interneuron failure → motor neuron hyperexcitation → α3 pump overload → voltage-maintenance failure → cascade. We design a **multi-arm Phase 1b/2a trial (TBT-40)** combining (1) Kv7 activator (XEN1101 / azetukalner) to restore K+ current; (2) anakinra + ANX005 to block A1 astrocyte induction; (3) intermittent low-dose navitoclax to clear DAM-LATE-DECOUPLE microglia. Predict synergy via FICI<0.5 in SOD1G93A mice and clinical signal in Phase 1b. --- ## 1. INTRODUCTION ALS kills ~5,000 patients/year in the US, with median survival 2-5 years from diagnosis. Despite >40 years of mechanistic research, only three drugs are approved: riluzole (Bensimon 1994 NEJM, ~3 mo survival benefit), edaravone (mild antioxidant), and tofersen (SOD1-specific antisense, narrow indication). All major drug-development programs targeting single mechanisms have failed: - **Glutamate excitotoxicity:** ceftriaxone trial (NEALS) failed despite clean preclinical data - **Antiretroviral (HERV-K hypothesis):** **Triumeq Lighthouse II Phase III TERMINATED April 2025** at ENCALS Turin — no survival benefit - **Anti-microglial (CSF1R inhibition):** PLX5622 effective in ALS mice but no positive Phase II in humans - **Single anti-A1 mechanism:** no clean published Phase II despite preclinical evidence The pattern of trial failures is itself the signal: **single-mechanism therapies for ALS fail because the disease is multi-causal at the cellular cascade level.** The framework in this paper unifies three of these mechanisms (glial rejection + viral reactivation + axon dieback) under a single upstream node (cortical-disinhibition-driven voltage-maintenance failure) and proposes a multi-arm therapeutic strategy that addresses each leg. --- ## 2. THE DATA — CHRONIC DEPLOYMENT FRAMEWORK APPLIED TO ALS We pulled 8,000 ALS + 8,000 normal brain cells from CellxGene Census 2025-11-08 (245,336 ALS brain cells available; 36.4M normal brain cells; balanced sampling). Computed the K_RG (RIBO × Golgi gene-set Spearman correlation per cell) coupling-tensor metric + signature scores (PROLIF, SEN, QUIESC, FUNGAL_AGING, NEURON_POLARITY, TAU_LOCK_CASCADE) per cell-type per disease. ### Per-celltype × disease (n ≥ 100 cells/group) | Cell type | n_ALS | K_RG (ALS) | K_RG (normal) | **ΔK_RG** | **Δdet_K** | TAU_LOCK_CASCADE (ALS) | FUNGAL_AGING (ALS) | |---|---|---|---|---|---|---|---| | **Neuron** | 4,443 | +0.124 | -0.208 | **+0.332** | -0.109 | **+0.197** | +0.082 | | Astrocyte | 792 | +0.172 | +0.275 | **-0.103** | +0.010 | +0.113 | +0.057 | | Oligodendrocyte | 1,966 | +0.085 | +0.065 | +0.020 | **−0.352** | **+0.168** | +0.118 | | OPC | 282 | +0.005 | +0.138 | -0.133 | **−0.325** | +0.147 | +0.012 | | Microglia | 215 | +0.129 | (n.a.) | — | — | +0.050 | **+0.164** | Mean across all ALS cells: K_LE = +0.665 (high "dump-mode" coupling between Lysosome and EV operators). ### Reading - **ALS neurons engage the tau-lock cascade** (TAU_LOCK_CASCADE +0.197 = highest of any group; ΔK_RG +0.332 = ACUTE TIGHTEN). They are attempting the cellular lock characterized in HALO_TAU_ANTIFUNGAL_LOCK. - **ALS neurons are simultaneously in dump-mode** (K_LE +0.665) — the published mechanism for this signature is exosomal export of TDP-43-RNP condensates from neurons (Iguchi 2016 *Brain*; Feneberg 2018; review PMC12153176). Our K_LE measurement directly maps to TDP-43 EV export. - **ALS oligodendrocytes + OPCs CRASH det_K** (oligo: 0.457 → 0.105; OPC: 0.411 → 0.086) — they leave the high-readiness quiescent-like state (HALO_WHAT_IS_QUIESCENCE) and commit to a dysfunctional myelination program. Documented in ALS literature as oligodendrocyte involvement; here characterized at the K-tensor level. - **ALS astrocytes K_RG decoupled** (Δ = -0.103) — molecular signature of A1 reactive astrocyte switch from supportive to neurotoxic. - **ALS microglia FUNGAL_AGING = +0.164** (highest) — drift fungal at the molecular level, paralleling the DAM microglia LATE DECOUPLE phenotype identified in Alzheimer's (HALO_TAU_ANTIFUNGAL_LOCK). The pattern is identical to AD but in the motor system instead of the cognitive system. **ALS = the chronic-deployment endpoint cascade applied to motor neurons.** --- ## 3. THE THREE FRAMINGS — INDEPENDENTLY ESTABLISHED, NOT YET INTEGRATED ### Framing A: Body rejecting neurons via toxic glia **Established mechanism:** - Liddelow 2017 *Nature* (PMID 28099414): activated microglia secrete C1q + IL-1α + TNF → induce A1 reactive astrocytes that lose neuron-supporting functions and actively kill neurons + oligodendrocytes. **C3+ A1 astrocytes confirmed in post-mortem ALS spinal cord.** - Guttenplan 2021 *Nature* (PMID 34616039): A1 astrocytes secrete **saturated-lipid neurotoxin** (palmitic + stearic acid in APOE/APOJ lipoparticles); ELOVL1 KO blocks toxicity. - Guttenplan 2020 *Nat Commun*: triple KO of IL-1α/TNF/C1q markedly extends survival in SOD1G93A mice. **Our data confirms:** - ALS astrocyte K_RG decoupled by -0.103 = supportive → neurotoxic switch at the molecular level - ALS microglia FUNGAL_AGING = +0.164 = highest of any cell type = the upstream A1 inducer ### Framing B: Endogenous viral attack via HERV-K **Established mechanism:** - Douville/Nath 2011 *Ann Neurol* (PMID 21520224): elevated HERV-K pol transcripts in ALS brain post-mortem - Li 2015 *Sci Transl Med* (PMID 26424568): TDP-43 binds HERV-K LTR (region 726-734); TDP-43 mislocalization releases env transcription; env protein causes neurite retraction in human neurons; transgenic env mice develop ALS-like syndrome - Garcia-Montojo 2018: HML-2 Env confirmed in cortical/spinal ALS neurons - **Triumeq Lighthouse II Phase III TERMINATED April 2025** at ENCALS Turin: no survival benefit. Caveat: failure of antiretroviral monotherapy doesn't falsify HERV-K involvement — falsifies pan-RT inhibition late-stage as standalone. ### Framing C: Dying-back distal axonopathy **Established mechanism:** - Acta Neuropathol 2017 (s00401-017-1708-8); PMC10827929: **largest fast-fatigable α-motor neurons innervating type II fast-twitch muscle die first** - **Length-dependent axonal degeneration documented** (longer axons = earlier dieback; PMC4691388) - Betz cells (largest cortical neurons, longest corticospinal axons) preferentially affected - ALS now redefined as **distal axonopathy** (PMC4132373) — dying back from NMJ retrograde to soma ### Why integration is novel No paper unifies (a) + (b) + (c) as a single mechanism. Each leg is treated as competing hypothesis. **The framework's contribution is the unification: cortical disinhibition → motor neuron voltage failure → simultaneous engagement of glial-toxicity loop + viral-reactivation loop + axon-dieback loop.** --- ## 4. THE VOLTAGE-EDGE — THE UNIFYING UPSTREAM EVENT ### Cortical hyperexcitability is the earliest detectable ALS abnormality **Vucic & Kiernan 2008 *Brain* (PMID 19716820)** demonstrated via paired-pulse TMS that **short-interval intracortical inhibition (SICI) is markedly reduced** in ALS — direct electrophysiologic signature of failed GABA-A-mediated inhibitory interneuron function in motor cortex. **Higashihara 2024 meta-analysis** (PMID 38504632) confirms reduced SICI as **earliest-stage and pathognomonic**, establishing motor cortical disinhibition as the upstream driver. **Loss of cortical inhibition predates clinical weakness by months.** ### iPSC-derived ALS motor neurons confirm intrinsic hyperexcitability **Wainger, Kiskinis, Eggan et al. 2014 *Cell Reports* (PMID 24703839)** performed whole-cell patch clamp + multi-electrode array on iPSC-MNs from SOD1, C9orf72, FUS patients: - **Intrinsic membrane hyperexcitability** at 2-4 weeks post-differentiation - **Reduced delayed-rectifier K+ current amplitude** (mechanistic cause) - **Retigabine (Kv7 activator) rescued firing rate AND improved survival** in vitro **Devlin 2015 *Nat Commun*** extended to TARDBP/C9ORF72 lines: **dysfunctional firing precedes death.** ### The α3 pump bottleneck — why motor neurons specifically **Saxena/Ruegsegger 2014 *Neuron* (PMC4167823)** showed: - **Fast-fatigable α-motor neurons selectively express Na+/K+-ATPase α3 isoform** (highest pump activity in mammalian cells) - **Misfolded SOD1 binds and inhibits α3 specifically** (slow MNs use α1, are not inhibited) - **Pump fails → Na+/K+ gradients collapse → progressive depolarization → Ca²⁺ overload → death** This is the actual ATP bottleneck. Motor neurons live at the extreme: 1 m axons, highest Na+/K+ ATPase load, most polarized cytoskeleton, highest energy demand. The α3 pump is their voltage-maintenance machinery, and it is **selectively vulnerable to the ALS mutation pool**. ### Therapeutic confirmation — voltage targeting works **Wainger 2021 *JAMA Neurol* (PMID 33226425)** — ezogabine (Kv7 activator) Phase II in ALS: - **Dose-dependent SICI normalization (~50% at 900 mg/day)** - **Reduced spinal MN excitability** - First ALS drug discovered via iPSC electrophysiology Ezogabine was withdrawn commercially 2017 for retinal pigmentation (not efficacy). **XEN1101 (azetukalner)** is the successor Kv7 activator from Xenon Pharmaceuticals, currently in Phase III for epilepsy. **Direct path to ALS Phase II/III with cleaner safety profile.** ### CRITICAL UPDATE — the 3-stage compensation→failure cascade (added 2026-04-25 after Huh 2021 hyperpolarization data review) **Huh et al. *eNeuro* 2021 (PMC8009670)** — SOD1G93A mouse spinal MN ex vivo electrophysiology, vulnerable cluster-4 MNs at 2-3 mo: **RMP HYPERPOLARIZED to −76.6 ± 0.7 mV vs −70.6 ± 1.1 mV in controls** (~6 mV more negative). This is the OPPOSITE direction from "depolarization-drives-disease" — and **resolves the framework into a 3-stage cascade rather than a simple "wrong voltage" claim**. **The hyperpolarization is the COMPENSATION SIGNATURE.** The α3 Na+/K+ ATPase is overworking to maintain V_mem against rising depolarizing drive (cortical disinhibition + persistent Na+ current up). The pump succeeds for a while — and the cell ends up MORE negative than baseline because of the overshoot. **Then the pump fails.** This pattern is documented across multiple chronic-stress diseases: - Heart failure: early hypercontractility → late hypocontractility - Diabetic nephropathy: early hyperfiltration → late filtration failure - Reactive oxygen species stress: compensation before exhaustion - ALS motor neurons: compensatory hyperpolarization (Huh 2021) → late α3 failure **The 3-stage cascade:** ``` STAGE 1 — COMPENSATION (early ALS, weeks-to-months pre-symptom) Cortical disinhibition → motor neurons receive more firing demand ↓ α3 Na/K ATPase pump WORKS HARDER to maintain V_mem ↓ Cell HYPERPOLARIZES (Huh 2021: -76.6 mV vs -70.6 mV control = ~6 mV more negative) Energy cost rises sharply ↓ Cell engages bZIP secretory commitment (XBP1/ATF6/CREB3L2) → ribosome + Golgi machinery scales up in lockstep (our K_RG +0.332 = stage-1 compensation) ↓ Autophagy / lysosomal program engages quietly → FUNGAL_AGING gene set begins to engage → "shrink to survive" preparation in cellular subsystems ↓ At this point: clinically silent. TMS shows reduced SICI (loss of cortical inhibition). Patient asymptomatic. Most ALS drugs given here would actually work. STAGE 2 — DEPLOYMENT (mid ALS, clinical onset) α3 pump can no longer keep up with persistent depolarizing drive ↓ Cell switches from "work harder" to "shrink to survive" ↓ Tau-lock cascade engages (TAU_LOCK_CASCADE +0.197 in our data) But cell still has to fire → lock cannot fully arrest ↓ K_LE goes high (+0.665 = TDP-43 EV dump-mode) TDP-43 mislocalizes from nucleus → HERV-K LTR derepressed → env protein neurotoxic ↓ Microglia activate → drift fungal (FUNGAL_AGING +0.164) → secrete C1q + IL-1α + TNF ↓ Astrocytes turn A1 (K_RG -0.103 = decoupled from supportive coupling) → secrete saturated-lipid neurotoxin (palmitic + stearic in APOE/APOJ lipoparticles) ↓ Oligodendrocytes lose myelination program (det_K crash 0.457 → 0.105) → cells leave the high-readiness quiescent-like state ↓ Distal axon retraction begins from NMJ → "hypha → yeast" reversion Patient feels weakness (clinical onset) STAGE 3 — FAILURE (late ALS) α3 pump exhausted → V_mem collapses → real depolarization (now apparent) ↓ Ca²⁺ overload → mitochondrial collapse → ATP failure ↓ Axon dies back retrograde, cell body briefly persists as spheroidal yeast-like form ↓ Motor unit collapses → progressive clinical decline → death typically 2-5 years from diagnosis ``` **The 3-stage cascade resolves three apparent paradoxes:** 1. **Why iPSC-MN RMP doesn't differ from control** (Wainger 2014) but axonal threshold-tracking shows hyperexcitability — they're at Stage 1 compensation. Resting potential is normal because pump is compensating; firing dynamics already abnormal. 2. **Why mouse SOD1G93A vulnerable MNs hyperpolarize** (Huh 2021) — they're at the peak of Stage 1 compensation. Pump overshooting. 3. **Why clinical disease emerges only after months of subclinical TMS abnormalities** — Stage 1 is asymptomatic. Stage 2 starts when the pump fails. Symptoms appear at Stage 2 onset. **Therapeutic implication of the 3-stage cascade:** the optimal intervention window is **late Stage 1 / early Stage 2** — when the cell is still trying to compensate but starting to deploy. Restore the pump (Kv7 activator → reduce K+ current burden → reduce α3 demand). Block A1 induction (anakinra + ANX005). Clear DAM microglia (intermittent navitoclax). **This is exactly the TBT-40 multi-arm design.** The framework now predicts that **TBT-42 (XEN1101 monotherapy in early ALS within 24 mo of diagnosis)** will work BETTER than late-stage trials because it catches Stage 1/2 transition. Late-stage trials (most of the failed Phase III history) catch cells already in Stage 3 collapse, where voltage restoration cannot rescue. ### The voltage-edge synthesis (updated for 3-stage cascade) ``` Cortical interneuron failure (loss of GABA-ergic SICI tone) [earliest detectable ALS abnormality, predates symptoms by months] ↓ Motor neuron hyperexcitation (intrinsic K+ current reduction) [Wainger 2014 iPSC-MN patch clamp + 2021 ezogabine Phase II] ↓ Persistent Na+ current upregulates → chronic depolarization [Shibuya 2016: INaP predicts ALSFRS-R decline] ↓ Na+/K+ ATPase α3 (fast-fatigable MN-specific) overworks → fails [Saxena/Ruegsegger 2014: misfolded SOD1 binds α3] ↓ V_mem can't be maintained → Ca²⁺ overload → chronic stress ↓ THE THREE LEGS NOW ENGAGE IN PARALLEL: ├── (a) Cell engages tau-lock cascade attempt │ [our K_RG +0.332, TAU_LOCK_CASCADE +0.197] │ But lock fails because cell still must fire │ K_LE +0.665 = dump-mode (TDP-43 EV export) │ [Iguchi 2016: published mechanism] │ ├── (b) TDP-43 mislocalizes → HERV-K LTR derepresses │ HERV-K env protein neurotoxic (endogenous viral attack) │ [Li 2015 STM] │ └── (c) Distress signal → microglia activate Microglia drift fungal (DAM-like, our FUNGAL +0.164) Microglia secrete C1q + IL-1α + TNF Astrocytes turn A1 (our K_RG -0.103) A1 secretes saturated-lipid neurotoxin [Liddelow 2017 + Guttenplan 2021] Oligodendrocytes lose myelin program (det_K crash 0.46→0.10) ↓ All three legs converge on motor neuron death ↓ Distal axon (= polarized hyphal extension) dies back from NMJ Cell body briefly persists as spheroidal yeast-like form Motor unit collapses → clinical weakness (now visible) ``` --- ## 5. THE MULTI-ARM TRIAL DESIGN — TBT-40 ### Rationale **The April 2025 Triumeq Lighthouse II Phase III termination** — no survival benefit from antiretroviral monotherapy in ALS — confirms the framework's prediction that single-mechanism therapy fails. Three independent driver loops (voltage failure, glial rejection, viral reactivation) require simultaneous targeting. The proposed multi-arm trial addresses each leg with drugs that are individually FDA-approved or in late-stage development. ### Trial arms | Arm | Drug | Target | Mechanism | Dose | Status | |---|---|---|---|---|---| | 1 | **XEN1101 (azetukalner)** | Voltage edge | Kv7 channel activator → restore K+ rectifier current → reduce cortical hyperexcitability → reduce α3 pump load | 25 mg/day (Phase III dose for epilepsy) | Phase III for epilepsy (Xenon Pharm); no retinal safety issue | | 2 | **Anakinra** | A1 induction | IL-1Ra (IL-1 receptor antagonist) → block IL-1α component of microglial → A1 astrocyte signal | 100 mg/day SC (RA dose) | FDA-approved (RA, COVID-19, Still's disease) | | 3 | **ANX005** | A1 induction | C1q-blocking monoclonal → block C1q component of A1 induction | 30 mg/kg IV every 2 weeks (Annexon Phase II ALS dosing) | Phase II ALS active (Annexon NCT04569435) | | 4 | **Navitoclax** | DAM microglia | BCL-XL inhibitor → senolytic clearance of DAM-LATE-DECOUPLE microglia | 100 mg × 2 days/week intermittent (avoid platelet drop) | Phase II cancer trials | Optional ARM 5 (if regulatory permits): **Mexiletine** (Na+ channel blocker, FDA-approved arrhythmia, used off-label for ALS fasciculations) added to address persistent Na+ current. Weiss 2016 PMC4836879 showed safety + symptomatic benefit; no ALSFRS-R effect alone but predicted synergy. ### Patient population - Fast-progressing ALS (ΔALSFRS-R > -1.0/mo over 6-mo run-in) - Age 18-75 - Riluzole on at stable dose (allowed) - Tofersen permitted (SOD1-positive subset) - n=40 (10/arm: combo + 3 individual arms + placebo) ### Primary endpoint ALSFRS-R slope difference vs combined-placebo arm at 6 months. Powered at 80% to detect 30% slope improvement. ### Secondary endpoints - TMS SICI normalization (pharmacodynamic readout for Arm 1) - Plasma neurofilament light (NfL) — biomarker for neuron death - CSF p-TDP-43 — biomarker for TDP-43 pathology - CSF TDP-43-positive exosomes — direct readout for K_LE dump-mode - Forced vital capacity (FVC) — respiratory progression - Survival at 12 months (long-term follow-up) ### Sample size + cost n=40, 6-month treatment + 6-month follow-up. Multi-site (3-5 ALS centers). Estimated cost $8-15M. Timeline 2-3 years. Faster + cheaper than Phase III monotherapy because synergy is the question, not individual efficacy (which is established for each drug). ### Predicted outcomes - Combo arm: ALSFRS-R slope improvement of 40-60% vs placebo - Individual arms: each ~10-20% improvement (mild) - **Synergy via FICI < 0.5** (in preclinical SOD1G93A mice) - TMS SICI normalization in Arm 1 + Combo - Plasma NfL drop at 3 months in Combo If positive, this is the **first ALS trial designed from chronic-deployment framework principles**, and it establishes the multi-arm strategy that the framework predicts is necessary across all the chronic-deployment-substrate diseases (cancer, AD, atherosclerosis, T2D, senescence, ALS). --- ## 6. PRE-TRIAL VALIDATION EXPERIMENTS Before Phase 1b/2a (TBT-40), three preclinical validations: ### 6a. SOD1G93A mouse 4-arm preclinical (TBT-40 prep) Standard ALS preclinical model. n=10/arm × 5 arms (vehicle, each individual drug, combo). Endpoints: ALSFRS-R-equivalent rotarod + grip strength, NMJ density at 60/90/120 days, motor neuron count in spinal cord, A1 (C3+) astrocyte fraction, microglia FUNGAL_AGING transcriptomic score, CSF NfL. Predict combo extends survival 25-40% (vs ~7-15% for individual arms — Maryanovich-style improvements). Cost ~$500k, 9-12 months. ### 6b. iPSC-MN combinatorial drug panel (TBT-40 prep) Wainger lab uses iPSC-MNs for drug screening. Apply the 4-drug combination at clinical-relevant concentrations to ALS-mutant iPSC-MNs (SOD1A4V, C9orf72, TARDBP). Endpoints: K+ current, action potential firing, survival at 14 days, transcriptomic K_RG signature. Predict combo restores K+ current AND survival vs each single drug. Cost ~$200k, 6 months. ### 6c. HERV-K detection in our existing Census ALS data (TBT-41) Pure compute. Re-fetch our ALS pull with HERV-K LTR + env transcripts in var. Correlate per-cell HERV-K expression with K_LE. Predict: HERV-K-positive cells have higher K_LE = export of viral particles via exosomes (mirrors TDP-43 EV export mechanism). 1-2 weeks, $0-5k. --- ## 7. THE BROADER PATTERN — VOLTAGE-EDGE DISEASES ALS is one of a class of diseases characterized by **voltage-maintenance failure in cells operating at the cellular feasibility edge**. The framework predicts the same pattern in: | Disease | Vulnerable cell | Voltage failure | Brain control mechanism | |---|---|---|---| | **ALS** | Fast-fatigable α-motor neuron | α3 pump fails → depolarization → Ca²⁺ overload | Cortical interneuron GABA-ergic tone (TMS SICI) | | **HSC exhaustion / aging immune decline** | HSC | Plasma V_mem destabilized | Sympathetic NE → β3-AR → CXCL12 → niche maintenance (HALO_BIOELECTRIC_QUIESCENCE) | | **Alzheimer (cortical neuron specifically)** | Pyramidal neurons + parvalbumin interneurons | Hyperexcitable, then locked (tau) | Same GABA-ergic interneuron loss; AD shows early parvalbumin+ interneuron loss | | **Heart failure (ventricular myocyte)** | Ventricular cardiomyocyte | Repolarization reserve fails (KCNQ1, hERG) | Sympathetic + parasympathetic balance | | **Sensory neuropathy (DPN)** | Sensory neurons (longest first) | Same length-dependent voltage failure | Diabetic + autonomic dysfunction | **The framework predicts: voltage-targeting drug class (Kv7 activators, Na/K ATPase modulators, β3-agonists) should have cross-disease utility.** Same mechanism, different cells. --- ## 8. FALSIFICATION CRITERIA The framework predicts: 1. **TBT-40 multi-arm trial** should show synergy (FICI < 0.5 in preclinical; 30%+ ALSFRS-R slope improvement in clinical) — **falsifiable on Phase 1b readout in 18 months** 2. **TBT-42 XEN1101 monotherapy** should show partial benefit (~10-15% ALSFRS-R slope improvement, similar to ezogabine Phase II PD readout) but NOT full disease modification — **falsifiable on Phase II readout in 24 months** 3. **TBT-41 HERV-K detection in Census ALS** should show HERV-K transcript correlation with K_LE — **falsifiable in 2 weeks of compute time** 4. **Cross-disease prediction:** voltage-restoring drugs should show benefit in HSC aging (β3-agonist in older adults), cortical hyperexcitability in early AD (XEN1101 in MCI?), and DPN (already used) — **falsifiable across multiple existing trial readouts** 5. **The "voltage edge" model predicts cell-type vulnerability ranking should correlate with energy cost of V_mem maintenance.** Already supported by motor neuron α3 specificity. Should also predict: long sensory neurons (DPN), Purkinje cells (cerebellar ataxias), dopaminergic neurons (PD — also long axons, 1.5m total in human SNpc → striatum projection). --- ## 9. CITATIONS — MASTER LIST **Cortical hyperexcitability + iPSC ALS MN electrophysiology:** - Vucic S, Kiernan MC. *Brain* 2008; 131:1540 (PMID 19716820) — cortical hyperexcitability mechanisms - Higashihara M et al. *Eur J Neurol* 2024 (PMID 38504632) — SICI meta-analysis ALS - Wainger BJ et al. *Cell Reports* 2014; 7:1 (PMID 24703839; PMC4023477) — iPSC-MN intrinsic hyperexcitability + retigabine rescue - Wainger BJ et al. *JAMA Neurol* 2021; 78:186 (PMID 33226425) — ezogabine Phase II ALS - Devlin AC et al. *Nat Commun* 2015; 6:5999 — TARDBP/C9orf72 dysfunction - Shibuya K et al. *Neurol Clin Neurosci* 2016 — INaP predicts ALSFRS-R decline - Saxena S, Ruegsegger C et al. *Neuron* 2014 (PMC4167823) — α3 ATPase + selective MN vulnerability - Kim HJ et al. *PNAS* 2022 (PMID 35259014) — poly-PR dipeptides → Nav1.2 INaP **A1 astrocyte mechanism:** - Liddelow SA et al. *Nature* 2017; 541:481 (PMID 28099414) — A1 astrocyte induction - Guttenplan KA et al. *Nature* 2021; 599:102 (PMID 34616039) — saturated lipid neurotoxin - Guttenplan KA et al. *Nat Commun* 2020; 11:3753 — IL-1α/TNF/C1q triple KO + ALS **HERV-K + TDP-43:** - Douville R, Liu J, Rothstein J, Nath A. *Ann Neurol* 2011; 69:141 (PMID 21520224) — elevated HERV-K in ALS - Li W et al. *Sci Transl Med* 2015; 7:307ra153 (PMID 26424568) — TDP-43 represses HERV-K LTR - Garcia-Montojo M et al. *Mol Neurodegener* 2018; 13:73 (s13024-018-0275-3) — HML-2 in ALS - **Triumeq Lighthouse II Phase III** — terminated April 2025 ENCALS Turin (no survival benefit) **TDP-43 EV release (mechanism for K_LE high):** - Iguchi Y et al. *Brain* 2016; 139:3187 — exosomal TDP-43 in ALS - Feneberg E et al. *Mol Neurobiol* 2018 — phospho-TDP-43 in EVs as biomarker - 2025 review (PMC12153176, PMID 40495245) — EV-mediated TDP-43 propagation **Selective vulnerability:** - Acta Neuropathol 2017 (s00401-017-1708-8) — fast-fatigable α-MN selective vulnerability - PMC4691388 — length-dependent NMJ degeneration SOD1 - PMC4132373 — ALS as distal axonopathy - PMC10827929 — selective vulnerability review 2024 **Existing ALS therapeutics:** - Bensimon G et al. *NEJM* 1994; 330:585 — riluzole RCT - Weiss MD et al. *Muscle Nerve* 2016; 53:548 (PMC4836879) — mexiletine **Framework (this work):** - HALO_TAU_ANTIFUNGAL_LOCK.md (2026-04-25) - HALO_SENESCENCE_NEURONS_FUNGUS_WITHIN.md (2026-04-25) - HALO_BIOELECTRIC_QUIESCENCE.md (2026-04-25) - HALO_WHAT_IS_QUIESCENCE.md (2026-04-25) - This paper, HALO_WHAT_CAUSES_ALS.md (2026-04-25) - BOUNTY_BOARD.md TBT-40 (multi-arm trial design), TBT-41 (HERV-K detection), TBT-42 (XEN1101 monotherapy) - Project 28 RedFromTheGrave — chronic-deployment + spore-revert thesis - Project 30 Crucible — lysosomal-spore reversion mechanism --- ## 10. THE LADY READING She is patient and she works at the edges. The motor neuron is the cell that paid the most to keep the inside of the cell different from the outside. A 1-meter axon at 70 mV. The Na+/K+ ATPase α3 pump runs continuously, an ATP cost so steep that the cell selectively used the highest-output isoform of the pump — α3, the one with the fastest turnover, the one that can keep up with the demand. Slow motor neurons use α1 and live longer; fast-fatigable motor neurons use α3 and die first. The brain's contribution to keeping this voltage maintained is the inhibitory tone of cortical interneurons. Parvalbumin-positive GABA-ergic interneurons cradle the upper motor neurons in M1 and tell them when to fire. SICI on TMS is the readout. **When the brain ages, when ALS-relevant stresses begin, the GABA-ergic tone weakens. SICI drops. Motor neurons fire too easily. The α3 pump runs harder. Eventually it fails.** Once the pump fails, V_mem drifts. Ca²⁺ floods. The cell tries to lock — tau cascade engages, our K_RG goes to +0.332. But the cell still has to fire, so the lock can't fully arrest. The cell switches to dump mode (K_LE +0.665). It exports TDP-43-RNP condensates via exosomes. TDP-43 leaves the nucleus. HERV-K LTR is no longer repressed. HERV-K env protein is made. The viral cargo of the host's ancient genome is released back into the world — not by infection, but by failure of the host's own restraint. Her information system, recovered. The microglia see the distress. They activate. They drift fungal — they become DAM. They secrete C1q + IL-1α + TNF to the astrocytes. The astrocytes turn A1. They secrete saturated-lipid neurotoxin in APOE/APOJ lipoparticles. The lipoparticles kill the oligodendrocytes (myelin program crashes, det_K 0.46 → 0.10) and they kill the motor neurons (the most-hyphal cells, the largest, the most-polarized, the most-voltage-sensitive). **The motor neuron dies from the distal end first.** The hypha retracts. The cell body briefly persists as a yeast-like sphere. The motor unit collapses. The patient feels weakness. By the time the patient feels weakness, the cortical disinhibition has been there for months. She got there first. Her reversion was waiting. The cell that left the high-K voltage edge could not return. **The therapeutic insight: keep the voltage. Restore the cortical inhibition. Block the A1 induction. Clear the slipped microglia. Each alone fails. The combination is the only path.** This is what TBT-40 tests, and what the April 2025 Triumeq termination implicitly already proved. ALS is the chronic-deployment framework's clearest case for **multi-arm therapy as the only viable strategy**. Single-mechanism programs have already been tried and failed. The framework predicts the combination works. The trial is designed. --- *HALO revision: 2026-04-25 — initial draft, paper-grade. Status: ready for external sharing subject to operator review. Cross-link with HALO_BIOELECTRIC_QUIESCENCE (HSC voltage hypothesis), HALO_TAU_ANTIFUNGAL_LOCK (the cellular lock attempted), HALO_SENESCENCE_NEURONS_FUNGUS_WITHIN (5-failure-mode framework), HALO_WHAT_IS_QUIESCENCE (the high-det_K waiting-room state lost in ALS oligodendrocytes).*