## TREM2 Mechanistic Analysis **Most Compelling: Stage-Selective Agonism (Early Amyloid Phase)** This hypothesis is most mechanistically compelling because it leverages the well-characterized DAM transition biology. TREM2 signaling via DAP12→SYK→PI3K-AKT-mTOR drives the metabolic switch to phagocytic DAM1 state in early AD, accelerating amyloid clearance before tau pathology cascades. **Evidence Supporting Agonism in Early AD:** - TREM2 R47H carriers show impaired DAM transition and accelerated plaque accumulation (Carmona et al., 2018) - AL002c agonist antibody restored microglial clustering around plaques in 5xFAD mice - TREM2 activation enhances PI3K-AKT-mTOR metabolic fitness enabling sustained phagocytosis - CSF sTREM2 levels correlate inversely with amyloid burden in MCI patients **Evidence Supporting Antagonism in Late-Stage Tauopathy:** - TREM2-high microglia in tau mice show increased TNF-α/IL-1β secretion damaging neurons (Leyns et al., 2019) - DAP12 signaling in late tau phase drives complement-mediated synapse elimination - TREM2 KO reduced tau-driven neurodegeneration in PS19 mice **Novel Insight (Stage-Selective + Signalosome Enhancement):** A biased TREM2 agonist targeting DAP12-SYK-PI3K WITHOUT activating the ERK-AP1 inflammatory arm could enable agonism across disease stages. **Top Biomarkers:** 1. CSF sTREM2 (microglial activation state) 2. Amyloid PET Centiloid (staging) 3. CSF p-tau181/t-tau ratio (tau propagation phase)

## Critical Evaluation **Stage-Selective Agonism:** - *Weak assumption:* CSF sTREM2 reliably predicts individual tissue microglial activation state - *Falsifying experiment:* Conditional TREM2 KO in DAM2 microglia during late tau phase - *Mouse model confound:* 5xFAD/PS19 kinetics don't match 10+ year human disease trajectory - *Trial risk:* Early vs late AD staging thresholds remain arbitrary **APOE4 Co-targeting:** - *Weak assumption:* APOE4 and TREM2 dysfunction operate through same lipid pathway simultaneously - *Falsifying experiment:* TREM2 agonism in APOE4-KI mice without APOE4 modifier — if efficacy maintained, co-targeting hypothesis is wrong - *Confound:* APOE4 affects BBB, not just microglia; unpredictable off-target effects **Biomarker-Guided Switching:** - *Weak assumption:* Transition point between agonism and antagonism is reliably detectable before irreversible neuronal loss - *Falsifying experiment:* Fixed-timing vs biomarker-guided switch in tau mice - *Critical confound:* TREM2 expressed in oligodendrocytes/astrocytes — microglial-selective modulation not currently achievable

## Clinical Feasibility Assessment **Most Actionable Now: Stage-Selective TREM2 Agonism (Early Amyloid Phase)** AL002c (Alector/AbbVie) completed Phase 1 with acceptable CNS penetration. Early-stage enrichment via amyloid PET (Centiloid >20) is immediately feasible. **Biomarker Strategy:** - Screening: Amyloid PET Centiloid 20-100 + APOE4 genotyping - Enrichment: Baseline CSF sTREM2 >2.0 ng/mL - Monitoring: CSF p-tau181 every 6 months - Stopping rule: p-tau181 doubling → observe or switch **Safety Risks:** 1. ARIA risk from accelerated amyloid clearance (as with anti-amyloid antibodies) 2. Peripheral effects: TREM2 in macrophages, osteoclasts — risk bone resorption abnormalities 3. Neuroinflammatory storm from over-activated phagocytosis 4. CNS delivery: IgG penetration 0.1-0.3%, high peripheral dosing needed **Timeline (Stage-Selective Agonism → Phase 2):** - Year 1-2: Biomarker threshold validation (ADNI, PREVENT-AD cohorts) - Year 2-3: Phase 1b dose-finding (amyloid PET-selected patients) - Year 3-5: Phase 2 RCT (n=200, 18-month primary endpoint) - Estimated cost: $45-80M through Phase 2

{"ranked_hypotheses": [{"rank": 1, "title": "Stage-Selective TREM2 Agonism", "composite_score": 0.72, "rationale": "Strongest mechanistic basis with active clinical development (AL002c Phase 1). Immediately actionable with amyloid PET staging.", "key_strength": "DAP12-PI3K-AKT mechanism well-characterized; existing clinical-stage agonist", "key_weakness": "ARIA risk and staging precision remain barriers"}, {"rank": 2, "title": "Stage-Specific Biomarker-Guided Switching", "composite_score": 0.68, "rationale": "Most intellectually complete framework resolving the agonism/antagonism paradox.", "key_strength": "Explicitly addresses stage-dependent controversy", "key_weakness": "Regulatory pathway for adaptive switching unprecedented"}, {"rank": 3, "title": "TREM2-APOE4 Co-targeting", "composite_score": 0.64, "rationale": "Strong synergistic rationale but requires concurrent dual-modality development.", "key_strength": "Addresses convergent lipid metabolism dysfunction", "key_weakness": "Combination complexity; APOE4-TREM2 interaction still preclinical"}, {"rank": 4, "title": "Soluble TREM2 (sTREM2) Therapeutic Mimic", "composite_score": 0.62, "rationale": "Innovative decoupling approach, partially characterized biology.", "key_strength": "Could avoid ARIA risk from full DAM activation", "key_weakness": "CNS half-life and dosing poorly characterized"}, {"rank": 5, "title": "TREM2-DAP12 Signalosome Enhancement", "composite_score": 0.60, "rationale": "Biased agonism concept enables stage-agnostic dosing.", "... [truncated]