Regulators' own account of the lecanemab-donanemab split reaches peer review

A year after this newspaper tracked Europe's reversal on both anti-amyloid antibodies, the officials who assessed them have published a formal, citable account of the reasoning behind two different regulatory paths to the same restricted approval.

A group of assessors from the European Medicines Agency's Committee for Medicinal Products for Human Use (CHMP), including its chair, has published a peer-reviewed account of how the agency weighed the benefits and risks of lecanemab and donanemab, the two anti-amyloid antibodies now authorised in the European Union for early Alzheimer's disease. The paper, a Personal View by Lorenzo Guizzaro and ten co-authors, appeared online on March 10, 2026 in The Lancet Regional Health - Europe (volume 63, article 101644), and moves the agency's reasoning from committee statements and press releases into the peer-reviewed literature, where it can be scrutinised, cited and challenged like any other clinical paper.

Both drugs followed the same formal route to market: an initial CHMP recommendation to refuse authorisation, an appeal by the manufacturer, and a reversal to a positive opinion after re-examination, first for lecanemab in November 2024 following a July 2024 refusal, then for donanemab in July 2025 following a March 2025 refusal. Both were ultimately authorised only for patients who are not homozygous carriers of the ApoE e4 gene variant, a group with a substantially higher rate of amyloid-related imaging abnormalities (ARIA), the brain swelling or bleeding that has been the central safety concern with both drugs. The authors report that when analysed with statistical methods designed to account for missing data after patients stopped treatment, lecanemab's effect over 18 months amounted to a difference of -0.535 points on the Clinical Dementia Rating Sum of Boxes scale (95% CI: -0.778 to -0.293) compared with placebo, alongside smaller but statistically significant differences on two other cognitive and functional scales.

The paper describes risk-minimisation measures, including a modified dosing and titration schedule for donanemab, as part of what shifted the agency's assessment from refusal to conditional approval on re-examination. It also states plainly that neither drug halts the disease and that both carry substantial risks, describing them instead as an early step toward more effective treatments. A separate rebuttal published two months later in the same journal, by clinicians led by Frank Jessen and Juan Fortea, agrees the pivotal trials were statistically robust but argues the outcome scales used to judge them were not designed to detect cognitive change in early-stage disease, a criticism the EMA's own horizon-scanning report has also raised.

Because the paper is written by the regulators who made the decisions, families and clinicians reading it are getting the agency's own reasoning in a citable, peer-reviewed form, not an independent audit of that reasoning; the published rebuttal in the same journal is one sign that debate over the underlying evidence has not closed.