On lecanemab, plasma biomarkers move in several directions at once

Design

Longitudinal cohort with untreated and amyloid-negative comparison groups

Population

197 patients on lecanemab (median age 73) plus large comparison cohorts

Primary outcome

Change in plasma Alzheimer's-related protein biomarkers over up to 2 years

Effect

34 of 130 markers changed (FDR <0.05) across four trajectories: normalising, overcorrecting, opposite, unchanged

A longitudinal cohort followed 197 patients with early symptomatic Alzheimer's disease receiving lecanemab in routine care, with large untreated and amyloid-negative comparison groups, tracking a panel of 130 plasma proteins over up to two years.

Thirty-four markers changed significantly and fell into four patterns: normalising towards healthy levels (phosphorylated tau 217, MAPT, GFAP), overcorrecting past them (amyloid-beta peptides), moving further from normal (inflammatory and synaptic markers such as TREM2, NPTXR, NRGN), and unchanged (neurofilament light, APOE). The divergence suggests treatment effects beyond amyloid clearance, including an immune response to plaques.

The practical implication is forward-looking: some markers may eventually monitor treatment response and others inform prognosis, but these are associations in a single-centre cohort and none is yet a validated response measure. It does not yet change how lecanemab is dosed or monitored.

• Of 130 plasma proteins, 34 changed significantly during lecanemab treatment.
• p-tau217, MAPT and GFAP partially normalised; amyloid-beta overcorrected.
• Inflammatory and synaptic markers (TREM2, NPTXR, NRGN) moved further from normal.
• Different markers may suit monitoring response versus prognosis, pending validation.
• This does not yet change lecanemab dosing or monitoring.

Why it matters

It previews how anti-amyloid therapy might be monitored, and cautions that biochemical change is not the same as clinical benefit.

Don't overread it

A single-centre observational cohort; biomarker changes are not shown to track cognitive outcomes and are not ready for clinical decisions.

The statistics, in plain English

These are observational biomarker trajectories, not clinical outcomes; a marker that normalises biochemically has not been shown to predict cognitive benefit, which is why validation is still needed.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.