Neurodegeneration

Neurodegeneration is not diffuse decay but a selective process: particular neuronal populations die while their neighbours survive, which is why these diseases have distinct clinical signatures. Most are proteinopathies, defined by a misfolded protein that aggregates and spreads along connected pathways, amyloid-beta and tau in Alzheimer's disease, alpha-synuclein in Parkinson's disease and Lewy body dementia, TDP-43 in most amyotrophic lateral sclerosis. Synaptic loss generally precedes cell death, and pathology precedes symptoms by many years.

The therapeutic record is mostly symptomatic. Levodopa transformed treatment of Parkinsonian motor signs without altering the loss of nigral neurons, and cholinesterase inhibitors and memantine do the same in Alzheimer's disease. The anti-amyloid antibodies lecanemab and donanemab, approved by the FDA in 2023 and 2024, are the first agents with a claim to modification, and they show the difficulty precisely: they clear amyloid from the brain almost completely on imaging while slowing decline on clinical scales by a small margin, with amyloid-related imaging abnormalities requiring monitoring.

The governing distinction is symptomatic versus disease-modifying. A drug that improves scores immediately and loses the difference on withdrawal is symptomatic; modification requires a change in the slope of decline, sustained through a delayed-start or randomised-withdrawal design.

Two inferences fail routinely. Neuroprotective and regenerative claims are made on the strength of a cell-viability assay measuring survival of a cultured line under an artificial insult, which predicts nothing about a human brain. And a biomarker that moves is treated as a benefit delivered, when the antibody experience shows near-total removal of a target pathology coexisting with a clinical effect at the edge of detectability.