---
title: Amyloid hypothesis
type: concept
last_updated: 2026-04-25
sources:
  - raw/papers/amyloid_hypothesis.pdf
  - raw/papers/biological_definition_of_AD.pdf
status: settled
---

# Amyloid hypothesis

The dominant pathophysiological model: Aβ accumulation is the upstream driver of AD, triggering tau pathology, neurodegeneration, and cognitive decline.

## Cascade (simplified)

`Aβ42 production / clearance imbalance → plaques → tau hyperphosphorylation → tangles → neuroinflammation → synaptic loss → neurodegeneration → cognitive decline`

Source: [`raw/papers/amyloid_hypothesis.pdf`](../../raw/papers/amyloid_hypothesis.pdf).

## Critique / open issues

- Aβ-targeting therapies (Lecanemab, Donanemab) reduce plaque burden but show only modest clinical benefit, and only in MCI / mild AD. This is consistent with — but does not prove — the hypothesis.
- Some patients with high amyloid burden remain cognitively normal; some with rapid decline have only modest amyloid. Tau extent (T2) and neurodegeneration (N) are stronger correlates of *symptom severity* than A alone.
- The 2024 biological definition (Jack et al.) accepts the cascade implicitly by making A+ a necessary condition, but stages by T2 extent — implicitly de-prioritizing A as a *progression* marker once the disease is established.

## Implications for this project

- The wearable signals correlate with the *symptomatic / neurodegeneration* tail of the cascade, not the upstream amyloid trigger.
- The system is therefore best framed as a **symptom-tracking and stage-monitoring** platform that complements (not replaces) biomarker testing.
- It is most useful in the SCD→MCI window where biomarkers may already be abnormal but standard scales are not yet sensitive.

## See also

- [AT(N) framework](atn_framework.md), [AD staging](ad_staging.md).