Test Preregistration Analysis

Test Preregistration: Therapeutic Hypotheses in Neurodegeneration

Hypothesis 1: Microglial TREM2 Activation as Therapeutic Target in Alzheimer's Disease

Title: TREM2 agonism will reduce amyloid-associated neurotoxicity by promoting microglial phagocytosis and metabolic reprogramming

Mechanism: TREM2 (Triggering Receptor Expressed on Myeloid Cells 2) is a microglia-specific receptor essential for microglial response to amyloid plaques. Agonist-mediated activation enhances amyloid clearance, reduces dystrophic neurites, and shifts microglia from a disease-associated (DAM) to homeostatic state.

Target Gene/Protein/Pathway: TREM2 → DAP12 → SYK signaling → metabolic reprogramming (glycolysis shift) → enhanced phagocytosis

Supporting Evidence:

• TREM2 R47H variant increases AD risk ~3-fold (PMID 24121985)
• TREM2-deficient mice show impaired microglial enclosure of amyloid plaques (PMID 29548884)
• Human PET imaging: TREM2 expression correlates with amyloid burden (PMID 31253634)
• TREM2 agonistic antibody (AL002c) shows efficacy in 5xFAD mice (PMID 32109293)

• In vivo: Amyloid PET (¹⁸F-AZD4694) reduction
• Ex vivo: Iba1+ cell coverage of plaques (immunohistochemistry)
• Behavioral: Morris water maze consolidation
• Molecular: RNAseq of sorted CD11b+ microglia (DAM signature genes)

Hypothesis 2: NLRP3 Inflammasome Inhibition Attenuates Neurodegeneration via IL-1β Suppression

Title: Chronic NLRP3 inhibition will prevent synapse loss in early-stage Alzheimer's disease through reduction of interleukin-1β mediated spine pruning

Mechanism: NLRP3 inflammasome activation in microglia releases active IL-1β, which drives synapse elimination through complement cascade upregulation (C1q, C3). Inhibition blocks this pathway, preserving synaptic density.

Target Gene/Protein/Pathway: NLRP3 → ASC → CASP1 → IL-1β → C1q/C3 upregulation → complement-mediated spine pruning

Supporting Evidence:

• NLRP3 KO mice protected from cognitive deficits in 3xTg AD model (PMID 26675736)
• IL-1β infusion causes spine loss in healthy mice (PMID 24489802)
• MCC950 (NLRP3 inhibitor) crosses BBB and reduces neuroinflammation (PMID 30638571)
• Active caspase-1 observed in AD human brains (PMID 12408820)

• Behavioral: Novel object recognition (episodic memory)
• Biochemical: Soluble Aβ42 (ELISA), mature IL-1β (meso scale)
• Morphological: Cortical dendritic spine density (Golgi-Cox)
• Electrophysiology: Long-term potentiation in hippocampal slices

Hypothesis 3: TFEB-Mediated Autophagy Enhancement for α-Synuclein Clearance

Title: TFEB nuclear translocation will reduce α-synuclein pathology by upregulating autophagosome-lysosome pathway genes

Mechanism: Transcription Factor EB (TFEB) is master regulator of lysosomal biogenesis. Forced nuclear localization using AAV9-TFEB(S211A) (phosphorylation-deficient mutant) will induce transcription of autophagy-lysosomal genes, enhancing clearance of toxic α-synuclein oligomers.

Target Gene/Protein/Pathway: mTORC1 inhibition → TFEB dephosphorylation → nuclear translocation → CLEA (autophagy/lysosome) gene transcription

Supporting Evidence:

• TFEB overexpression reduces α-synuclein aggregation in BAC transgenic mice (PMID 24185427)
• mTORC1 inhibition (rapamycin) enhances α-synuclein clearance (PMID 17442627)
• PBAE nanoparticle-mediated TFEB delivery reduces polyQ aggregation (PMID 29890108)
• Autophagy-lysosome pathway genes downregulated in PD substantia nigra (PMID 28794040)

• Behavioral: Rotarod, cylinder test (motor function)
• Biochemical: p62/SQSTM1, LC3-II (autophagy markers), α-synuclein pSer129 (ELISA)
• Histological: p62+ aggregates, NeuN+ neuron count in substantia nigra
• Survival endpoint: Incidence of paralysis

Hypothesis 4: SIRT1 Activation as Mitochondrial Protective Strategy in ALS/FTD

Title: SIRT1 activation will suppress C9orf72-mediated neurodegeneration by restoring mitochondrial homeostasis and reducing DPR accumulation

Mechanism: C9orf72 repeat expansions produce toxic gain-of-function through bidirectionally transcribed RNA foci and five dipeptide repeat (DPR) proteins. SIRT1 activation promotes mitophagy (via PINK1/Parkin), reduces oxidative stress, and has been shown to decrease DPR levels in cellular models.

Target Gene/Protein/Pathway: SIRT1 → PGC-1α deacetylation → mitochondrial biogenesis; SIRT1 → LC3 deacetylation → autophagy flux

Supporting Evidence:

• SIRT1 activator (SRT2104) extends survival in TDP-43 mice (PMID 26805578)
• C9orf72 KO mice show mitochondrial dysfunction and lysosomal accumulation (PMID 25908856)
• SIRT1 overexpression reduces poly-GR toxicity in Drosophila (PMID 31278169)
• Resveratrol (SIRT1 activator) improves mitochondrial function in patient-derived motor neurons (PMID 29469839)

• Molecular: Expression of mitochondrial complex subunits (Complex I-V western blot), PGC-1α acetylation state
• Behavioral: Grip strength, rotarod latency, nesting behavior
• Histological: Motor neuron count (ChAT+ in ventral horn), Iba1+ gliosis
• Survival: Cumulative survival analysis (log-rank test)
• Tissue: DPR burden (poly-GA, poly-GR immunohistochemistry)

Hypothesis 5: CDK5 Dysregulation in Tau Pathology Propagation

Title: CDK5 inhibition will block activity-dependent tau propagation by preventing phosphorylation of tau at S202/T205 and reducing exosome secretion

Mechanism: Neuronal activity induces CDK5-dependent tau phosphorylation and packaging into exosomes. CDK5 inhibition (using dinaciclib or peptide inhibitors) will prevent this loading and reduce trans-synaptic tau spreading.

Target Gene/Protein/Pathway: Neuronal activity → Ca²⁺ → p25/p35 → CDK5 activation → tau S202/T205 phosphorylation → exosome loading → extracellular spreading

Supporting Evidence:

• p25/CDK5 hyperactivity drives tau hyperphosphorylation in AD (PMID 15745994)
• Exosome-mediated tau spread confirmed in human CSF (PMID 29072881)
• Dinaciclib (CDK5 inhibitor) reduces tau phosphorylation in vivo (PMID 26593258)
• Tau propagation requires neuronal activity (PMID 26863191)

• Biochemical: p-tau S202/T205, total tau (Mab HT7), exosomal tau (NTA analysis)
• Anatomical: Dentate gyrus p-tau burden (Stereology)
• Cognitive: Contextual fear conditioning
• In vitro: Primary neuron uptake of exosomes from treated vs. control mice

Hypothesis 6: NRF2 Antioxidant Pathway as Common Mechanism Across Neurodegeneration

Title: NRF2 activation will provide neuroprotection across ALS, AD, and PD through restoration of glutathione-dependent redox buffering

Mechanism: NRF2 (NF-E2-related factor 2) is master regulator of antioxidant response elements (ARE). Genetic or pharmacologic NRF2 activation (CDDO-EA, sulforaphane) will upregulate NQO1, HO-1, GCLM, and GSTA1, restoring redox homeostasis impaired in all major neurodegenerative diseases.

Target Gene/Protein/Pathway: Keap1 oxidation → NRF2 release → nuclear translocation → ARE gene transcription → NQO1, HO-1, GCLM, GCLC, GSTA1

Supporting Evidence:

• NRF2 activation protects in MPTP mouse model of PD (PMID 29104108)
• NRF2 activator (CDDO-TFEA) crosses BBB and extends ALS mouse survival (PMID 28467821)
• NRF2 target gene polymorphisms associated with AD risk (PMID 20819947)
• Post-mortem AD brain shows impaired NRF2 nuclear localization (PMID 23563891)

• Motor: Rotarod, grip strength
• Biochemical: GCLM expression, GSH/GSSG ratio, 4-HNE adducts
• Survival (ALS model only)
• RNAseq: ARE gene signature in motor cortex/ventral spinal cord

Hypothesis 7: Synaptic NMDAR Subtype Targeting to Prevent Excitotoxicity

Title: GluN2A-containing NMDA receptor activation will prevent synapse loss better than broad NMDAR antagonism in Alzheimer's disease

Mechanism: Excessive extrasynaptic NMDAR (GluN2B-containing) activation drives CREB shutoff and excitotoxicity, while synaptic NMDAR (GluN2A-containing) promotes survival signaling. Selective GluN2A activation or GluN2B inhibition will shift balance toward neuroprotection.

Target Gene/Protein/Pathway: Synaptic NMDAR (GluN2A) → Ca²⁺ → CaMKII → CREB → BDNF expression → synaptic stability; Extrasynaptic NMDAR (GluN2B) → PP1 → pCREB shutoff

Supporting Evidence:

• Extrasynaptic NMDAR overactivation triggers synaptotoxicity in AD models (PMID 20371869)
• Ifenprodil (GluN2B antagonist) prevents Aβ-induced dendritic spine loss (PMID 15229342)
• TCN-201 (GluN2A antagonist at synaptic sites) worsens outcome, suggesting differential compartment effects
• Memantine (partial NMDAR antagonist) preferentially blocks extrasynaptic receptors (PMID 17928558)

• Electrophysiology: CA1 LTP (fEPSP slope)
• Biochemical: pCREB S133, proBDNF/mature BDNF ratio, CaMKII autophosphorylation
• Behavioral: Contextual fear conditioning, Y-maze spontaneous alternation
• Histological: Synaptophysin+ puncta density (electron microscopy gold standard)

Summary Table

| Hypothesis | Target | Primary Disease | Confidence | Evidence Quality |
|------------|--------|-----------------|------------|------------------|
| 1 | TREM2 | AD | 0.78 | High (multi-species) |
| 2 | NLRP3/IL-1β | AD | 0.72 | Moderate |
| 3 | TFEB | PD/LBD | 0.70 | Moderate |
| 4 | SIRT1 | ALS/FTD | 0.68 | Moderate |
| 5 | CDK5 | AD/FTD | 0.74 | Moderate-High |
| 6 | NRF2 | All | 0.76 | High |
| 7 | NMDAR subtype | AD | 0.65 | Moderate | Preregistration timestamp: [DATE] | Analysis plan: Two-way ANOVA (genotype × treatment) with Bonferroni correction for behavioral and biochemical endpoints. Sample size: n=12/condition (80% power for d=0.8). Blinded scoring for all histological endpoints.

Skeptic's Evaluation: Neurodegeneration Therapeutic Hypotheses

Hypothesis 1: TREM2 Agonism

Confounds:

• 5xFAD mice have aggressive amyloid deposition; microglial dynamics differ in human AD where plaques form over decades
• TREM2 agonism may enhance phagocytosis of vulnerable synapses, not just plaques (PMID 30742032 showed microglial engulfment of excitatory terminals)
• AL002c is hypothetical—has this been independently replicated?

Falsifying experiment: Treat 12-month 5xFAD mice (established plaques) with AL002c. Hypothesis 1 predicts improved cognition and reduced plaque burden. Alternative prediction: No behavioral benefit despite increased microglial coverage (plaques persist).

Revised confidence: 0.65 (down from 0.78)

Hypothesis 2: NLRP3 Inhibition

Confounds:

• MCC950's BBB penetration in the cited paper (PMID 30638571) used 6-week chronic dosing in young mice; pharmacokinetics in symptomatic aged mice are unknown
• IL-1β infusion is acute; AD pathology develops over years—chronic IL-1β exposure may have different mechanisms

Falsifying experiment: Treat 12-month APP/PS1 mice with MCC950 after behavioral deficits are established. Hypothesis 2 predicts reversal of deficits. Alternative prediction: No behavioral rescue; detectable IL-1β in plasma but not CSF.

Revised confidence: 0.58 (down from 0.72)

Hypothesis 3: TFEB

Confounds:

• AAV9 delivery efficiency in aged mice varies significantly; 10⁹ genomic copies may be insufficient for therapeutic effect
• M83 mice develop α-syn pathology but not human LB disease; pathology is primarily axonal, not cellular
• mTORC1 inhibition (rapamycin) has pleiotropic effects—attributing effects solely to TFEB ignores systemic autophagy dysregulation

Revised confidence: 0.55 (down from 0.70)

Hypothesis 4: SIRT1 in C9orf72

Confounds:

• Drosophila models of C9orf72 are not validated for translational relevance to human disease
• SRT2104 is described but has poor oral bioavailability; pharmacokinetics in mice are poorly characterized
• C9orf72 KO mice show mitochondrial dysfunction and lysosomal accumulation—SIRT1 activation addresses only the mitochondrial component

Falsifying experiment: Measure DPR levels (poly-GA, poly-GR) via ELISA in cortex after SRT2104 treatment. Hypothesis 4 predicts reduction. Alternative prediction: DPR accumulation continues; survival benefit is due to general neuroprotection, not C9orf72-specific pathway.

Revised confidence: 0.52 (down from 0.68)

Hypothesis 5: CDK5 Inhibition

Confounds:

• Optogenetic 40Hz stimulation is highly non-physiological; may induce seizure-like activity in aged mice
• AAV-dnCDK5 expression needs independent validation (antibody confirmation, enzymatic assay)
• Exosomal tau spread in human AD is correlative; causality not established

Falsifying experiment: Use a selective CDK5 peptide inhibitor (CIP) instead of dnCDK5. Compare exosomal tau from treated vs. control mice in a primary neuron uptake assay. Alternative prediction: dnCDK5 reduces p-tau but exosomal tau unchanged—tau propagation uses non-exosomal mechanisms.

Revised confidence: 0.60 (down from 0.74)

Hypothesis 6: NRF2 Across Diseases

Confounds:

• CDDO-EA has off-target effects (PPARγ, NF-κB inhibition); attributing results solely to NRF2 is problematic
• NRF2 activation in post-mortem AD brains may be a compensatory response, not a driver of pathology—activating it pharmacologically may not add benefit

Falsifying experiment: Include NRF2 KO mice in each model. Hypothesis 6 predicts CDDO-EA fails in NRF2 KO mice. Alternative prediction: NRF2 KO mice show same protection—CDDO-EA acts via NRF2-independent pathways.

Revised confidence: 0.60 (down from 0.76)

Hypothesis 7: NMDAR Subtype Targeting

Confounds:

• Ifenprodil also binds sigma-1 receptors; neuroprotective effects may be sigma-1 mediated, not GluN2B-specific
• TCN-201 blocks synaptic GluN2A but has off-target effects at high doses

Falsifying experiment: Use genetically modified mice with conditional GluN2B deletion in forebrain excitatory neurons (not global KO). Alternative prediction: Synapse loss continues despite extrasynaptic blockade—compensatory mechanisms bypass the hypothesized pathway.

Revised confidence: 0.48 (down from 0.65)

Summary of Revised Confidence Scores

| Hypothesis | Original | Revised | Primary Issue |
|------------|----------|---------|--------------|
| TREM2 | 0.78 | 0.65 | R47H ≠ pharmacologic agonism; synapse engulfment risk |
| NLRP3 | 0.72 | 0.58 | Developmental compensation; BBB penetration uncertainty |
| TFEB | 0.70 | 0.55 | AAV delivery efficiency; non-physiological mechanism |
| SIRT1 | 0.68 | 0.52 | Pleiotropic targets; Drosophila model validity |
| CDK5 | 0.74 | 0.60 | Non-selective inhibitor; optogenetic confounds |
| NRF2 | 0.76 | 0.60 | Overly broad mechanism; terminal-stage intervention |
| NMDAR | 0.65 | 0.48 | Receptor compartmentalization unproven |

Across all hypotheses:

• Mouse model validity (aggressive transgene expression ≠ human disease)
• Therapeutic timing (pre-symptomatic vs. established pathology)
• Mechanistic specificity (pleiotropic pathways, off-target effects)

Domain Expert Assessment: Neurodegeneration Therapeutic Hypotheses

Overview

The Theorist presented 7 mechanistically-plausible hypotheses; the Skeptic's critiques substantially weakened most. Below I assess feasibility across five dimensions for each surviving hypothesis.

H1: TREM2 Agonism in AD

| Dimension | Assessment |
|-----------|------------|
| Druggability | Tractable — antibody therapeutics suit cell-surface receptors. AL002c (from Alector) already in Phase II for AD. Fc-mediated agonism is a proven mechanism. |
| Biomarkers/Models | Model concern: 5xFAD has aggressive transgene-driven amyloid deposition over weeks, unlike human AD's decades-long evolution. Consider App knock-in models (e.g., APP^NL-G-F) for better translation. Biomarkers: Plasma sTREM2, PET amyloid, CSF Iba1 + microglia RNA signatures are trackable. |
| Clinical Development | Timing window is critical: Microglial state transitions occur early in disease. By time of clinical diagnosis, microglia may already be "exhausted." biomarker-guided enrollment (TREM2 + amyloid PET positivity) essential. |
| Safety | The Skeptic's synapse engulfment concern is valid. DAM microglia can prune excitatory synapses (PMID 30742032). Pharmacologic agonism in humans risks enhancing microglia-mediated synapse loss. Requires synapse-density monitoring ( synaptic neurofilament in CSF) in trials. |
| Timeline/Cost | Phase I already complete (AL002c). Phase II requires 18-24 months, $15-25M. If results replicate Alector's Phase II signals (they were mixed), Phase III (300+ patients, 3 years, $80-120M) could begin ~2028-2029. |

Verdict: Most translation-ready. But enroll early-stage patients, monitor synaptic markers, and use knock-in models for preclinical validation.

H2: NLRP3 Inhibition via IL-1β Suppression

| Dimension | Assessment |
|-----------|------------|
| Druggability | Moderate. MCC950 is a research tool with undefined oral bioavailability and unknown chronic toxicology. Dapansutrile (OC000459) is the clinical-stage candidate — orally bioavailable, completed Phase II for gout (modest efficacy). CNS penetration remains the core problem. |
| Biomarkers/Models | BBB penetration in symptomatic aged mice needs independent replication. IL-1β in CSF vs. plasma dissociation is a known confound — don't rely on plasma levels. Complement cascade markers (C1q, C3) in CSF are downstream readouts. |
| Clinical Development | Already in inflammatory disease trials (advantage). Key regulatory question: can you approve a compound for CNS indication based on peripheral inflammation biomarkers? Likely requires CSF target engagement data. |
| Safety | Infection risk with chronic IL-1β blockade is real (historically increased TB reactivation with IL-1 inhibitors). Requires TB screening, careful infection monitoring. |
| Timeline/Cost | Re-purposing dapansutrile for AD: Phase IIa biomarker study (CSF IL-1β engagement) ~12-18 months, $8-12M. Phase IIb cognitive endpoint depends on Phase IIa results. Total to Phase III start: 4-6 years, $40-60M. |

Verdict: Viable but requires BBB-penetration validation. The field needs a CNS-optimized NLRP3 inhibitor with demonstrated CSF target engagement before committing to large AD trials.

H3: TFEB-Mediated Autophagy Enhancement

| Dimension | Assessment |
|-----------|------------|
| Druggability | Challenging. Gene therapy (AAV9-TFEB) is the proposed approach — this is a high-risk, high-cost modality for a first-in-class indication. Non-viral alternatives ( nanoparticle delivery of TFEB mRNA) are earlier-stage. |
| Biomarkers/Models | M83 mice model α-synucleinopathy but not Lewy body disease — axonal pathology predominates. For translational relevance, need models with confirmed LB-like pathology (e.g., Thy1-αSyn or patient-derived systems). Autophagy flux assays (tandem RFP-GFP-LC3) are essential for target engagement. |
| Clinical Development | AAV9 CNS delivery faces manufacturing scale-up challenges. $2-5M per patient for commercial gene therapy. Insurance/reimbursement for neurodegenerative indications unresolved. |
| Safety | mTORC1 inhibition causes metabolic effects ( immunosuppression, hyperlipidemia). TFEB overexpression may dysregulate lysosomal trafficking (potential worsening of lysosomal storage). |
| Timeline/Cost | Preclinical validation + IND-enabling studies: 3-4 years, $20-30M. Phase I (safety, dose escalation): 2 years, $15-20M. This is a 10+ year path to approval. |

Verdict: Too early for clinical development. Refine AAV delivery in appropriate models, establish autophagy flux biomarkers, demonstrate functional benefit before committing to this path.

H4: SIRT1 in C9orf72 ALS/FTD

| Dimension | Assessment |
|-----------|------------|
| Druggability | Moderate, but target specificity is problematic. SRT2104 is a first-generation SIRT1 activator with unclear selectivity vs. SIRT2/3. Next-gen activators (SIRT2104 analogs, SIRT1-activating compounds) need head-to-head comparison. |
| Biomarkers/Models | Drosophila C9orf72 models have limited translational value — DPR toxicity mechanisms differ from mammals. SOD1*G93A mice are the best-characterized ALS model but don't capture C9orf72-specific DPR pathology. Need C9-BAC or patient-derived motor neurons for validation. |
| Clinical Development | Resveratrol's failure in AD (PMID 24445164) is a warning signal. SIRT1 activation may have failed there due to insufficient CNS penetration or wrong disease stage. For ALS/FTD, earlier intervention (presymptomatic carriers) may be necessary. |
| Safety | SIRT1 affects metabolism ( glucose homeostasis), p53, FOXO pathways — broad activation carries cancer and metabolic risks with chronic dosing. |
| Timeline/Cost | SRT2104 is already in Phase I for psoriasis/diabetes — could be repurposed. ALS Phase II biomarker study: 18-24 months, $12-18M. Orphan designation (ALS) reduces development costs. |

Verdict: Marginally feasible. Resveratrol's failure demands caution, but the mechanism deserves testing with better tools (selective activators, C9-specific models, patient-derived cells).

H5: CDK5 Inhibition for Tau Propagation

| Dimension | Assessment |
|-----------|------------|
| Druggability | The Skeptic is correct: dinaciclib is not a selective CDK5 inhibitor. For clinical development, need selective CDK5 inhibitors (e.g., CSP-110/RO-890 or peptide inhibitors like CIP). Dinaciclib's CDK1/2/9 activity confounds interpretation. |
| Biomarkers/Models | P301S tau mice are standard. 40Hz optogenetic stimulation is non-physiological and may induce seizures — use chemogenetics (DREADDs) or enriched environment instead. Exosomal tau measurement requires standardization (NTA vs. ELISA, isolation protocol). |
| Clinical Development | Narrow therapeutic window is the key constraint. CDK5 is essential for synaptic plasticity and memory. Complete CDK5 inhibition will impair cognition. Partial/activity-dependent inhibition may be required. |
| Safety | Pan-CDK inhibitors (dinaciclib) cause cytopenias, GI toxicity. Selective CDK5 inhibitors have not been in human trials — unknown safety profile. |
| Timeline/Cost | Development of selective CDK5 inhibitor + IND-enabling studies: 3-4 years, $25-35M. Phase I safety (CNS penetration, cognitive monitoring): 2 years, $15-20M. |

Verdict: Mechanistically interesting but requires better tools. The narrow therapeutic window may be prohibitive if systemic CDK5 inhibition cannot achieve sufficient brain penetration without cognitive side effects.

H6: NRF2 Across ALS/AD/PD

| Dimension | Assessment |
|-----------|------------|
| Druggability | Well-established target with multiple clinical-stage compounds. CDDO-EA (bardoxolone methyl) has been in Phase III for CKD and COPD. Sulforaphane is in trials for autism and schizophrenia. BBB penetration varies by compound. |
| Biomarkers/Models | ARE gene signature (NQO1, HO-1, GCLM expression) in blood or tissue is measurable. Triple-disease approach is ambitious — each indication may require different timing and endpoint optimization. |
| Clinical Development | The "terminal oxidative stress" convergence hypothesis is biologically plausible but may be too late-stage. By the time NRF2 activation achieves meaningful effects, substantial neuronal loss has already occurred. Consider prevention trials in genetic at-risk populations (e.g., LRRK2 G2019S carriers for PD). |
| Safety | CDDO-EA has off-target effects (PPARγ, NF-κB) and caused cardiac events in some CKD trials. Sulforaphane has a favorable safety profile but modest potency. |
| Timeline/Cost | Repurposing existing NRF2 activators: Phase II biomarker study (ARE gene signature, GSH/GSSG ratio) in each indication: 12-18 months each, $10-15M. Cross-indication development reduces per-indication costs. |

Verdict: Most practical near-term opportunity. Multiple clinical-stage compounds exist. Focus on early-stage/presymptomatic populations to test the "oxidative stress convergence" hypothesis.

H7: NMDAR Subtype Targeting

| Dimension | Assessment |
|-----------|------------|
| Druggability | Moderate — ifenprodil is a known compound, but its sigma-1 receptor binding confounds mechanism attribution. GluN2A-selective positive modulators are harder to develop (agonists vs. positive allosteric modulators). |
| Biomarkers/Models | In vivo compartmentalization is technically challenging to confirm. Use genetically-encoded NMDAR sensors (iGluSnFR variants) or region-specific biochemistry. |
| Clinical Development | Memantine's modest efficacy in humans is the strongest evidence against this hypothesis. If memantine (partial extrasynaptic blocker) provides limited benefit, ifenprodil may not do much better. |
| Safety | NMDAR modulation carries excitotoxicity risk (over-inhibition can be as dangerous as over-activation). GluN2B deletion in forebrain excitatory neurons causes cognitive impairment in mice — confirms the narrow therapeutic window. |
| Timeline/Cost | Repurposing ifenprodil: Phase II cognitive endpoint in AD: 18-24 months, $15-20M. If results are negative, this program should be deprioritized. |

Verdict: Least translation-promising. Memantine's clinical reality argues against this mechanism. Only pursue if new selective GluN2A modulators become available.

Prioritization Summary

| Rank | Hypothesis | Rationale | Confidence | Key Next Step |
|------|------------|-----------|------------|---------------|
| 1 | H6 (NRF2) | Clinical-stage compounds, multiple indications, favorable risk profile | 0.60 | Phase II biomarker study in genetic at-risk populations |
| 2 | H1 (TREM2) | Most advanced (AL002c in Phase II), antibody therapeutics well-established | 0.65 | Monitor Alector's Phase II results; use knock-in models |
| 3 | H2 (NLRP3) | Repurposing opportunity, CNS penetration needs validation | 0.58 | Test CNS-penetrant NLRP3 inhibitors (dapansutrile analog) |
| 4 | H5 (CDK5) | Novel mechanism, requires selective inhibitor development | 0.60 | Develop selective CDK5 inhibitor, validate therapeutic window |
| 5 | H4 (SIRT1) | Repurposing possible, but resveratrol failure is concerning | 0.52 | Test with selective activator in C9-specific models |
| 6 | H3 (TFEB) | Gene therapy path, too early for clinical development | 0.55 | Establish autophagy biomarkers, non-viral delivery |
| 7 | H7 (NMDAR) | Memantine's failure argues against this mechanism | 0.48 | Deprioritize unless novel selective modulators emerge |

Bottom line: H6 (NRF2) and H1 (TREM2) warrant immediate clinical development. H3 (TFEB) and H7 (NMDAR) require either substantial preclinical work or deprioritization.