All News

Lecanemab at Five Years: Reading the Long-Term Evidence

By Sandy Martin•November 18, 2026

A company conference update reports that lecanemab’s benefit in early Alzheimer disease continued through five years. Published human extension data reach 48 months, helping patients and clinicians assess durability and treatment duration.

Eisai and Biogen have reported that the clinical benefit associated with lecanemab continued through five years in people treated for early Alzheimer disease. A report on the conference presentation says the companies presented the results from the CLARITY-AD long-term extension at the Clinical Trials on Alzheimer’s Disease meeting in Boston.

The update matters because lecanemab is intended to alter the course of disease rather than provide a short-lived improvement in symptoms. Evidence that treated patients continue to decline more slowly over several years would strengthen the argument that removing amyloid can produce a durable clinical effect.

The evidence stages should not, however, be treated as interchangeable. The five-year result is currently a company-reported conference update. The peer-reviewed extension data available in full reach 48 months, providing substantial context for the new presentation but not yet an independently appraisable account of its additional year of follow-up.

A five-year signal, with four years in print

The published open-label extension analysis included 1,734 participants who received lecanemab during either the original trial or its extension. It reported clinical outcomes through 48 months, a considerably longer period than the blinded trial’s primary 18-month comparison.

Using a matched cohort from the Alzheimer’s Disease Neuroimaging Initiative, or ADNI, as an external control, the researchers calculated a difference on the Clinical Dementia Rating–Sum of Boxes, known as CDR-SB, of 0.52 points at 18 months, 1.01 points at 36 months and 1.75 points at 48 months. CDR-SB combines assessments of cognition and everyday function, with higher scores indicating greater impairment.

In this setting, “benefit” does not mean that lecanemab restored abilities already lost to Alzheimer disease. It means that decline was slower in treated participants than in the comparison group. The widening difference through 48 months is consistent with an effect that accumulates as the disease progresses, provided the external comparison accurately represents how the treated participants would otherwise have fared.

The five-year presentation appears to carry that trajectory forward. Until the full analyses are published, however, its numerical results, statistical methods and participant follow-up cannot be assessed with the same detail as the 48-month paper.

The randomized foundation remains 18 months

CLARITY-AD began as a phase 3, randomized, double-blind trial involving 1,795 adults with early Alzheimer disease. Participants received intravenous lecanemab at 10 milligrams per kilogram every two weeks or placebo for 18 months. This randomized period provides the program’s strongest evidence that differences between groups were caused by treatment rather than by underlying differences among participants.

The trial population was narrowly defined: adults aged 50 to 90 with mild cognitive impairment due to Alzheimer disease or mild Alzheimer dementia, plus confirmed amyloid pathology. That definition remains important when interpreting longer follow-up. The evidence concerns treatment initiated at an early symptomatic stage; it does not automatically extend to people without symptoms or those with more advanced dementia.

Lecanemab is a monoclonal antibody directed against aggregated forms of amyloid-beta. By reducing amyloid deposits, it targets a defining part of Alzheimer pathology rather than directly stimulating memory or attention. A durable clinical separation would therefore support the proposed disease-modifying mechanism, although amyloid removal and measured clinical benefit are distinct outcomes.

How the extension changes the comparison

After the blinded phase, the study became open-label and all participants could receive lecanemab. There was no longer a concurrently randomized, untreated group. Longer-term analyses consequently depend on comparisons between participants who started at different times and on matched natural-history cohorts such as ADNI.

That approach is useful when maintaining a placebo group for years is impractical, but it is less resistant to bias than randomization. Matching can align groups on measured characteristics; it cannot ensure that they were equivalent in every factor affecting progression, clinical care or continued study participation. A review of the long-term evidence therefore describes these results as resting on delayed-start comparisons and external natural-history assumptions after the initial trial.

This does not make the extension uninformative. Following a large treated cohort for four years can reveal whether an early effect disappears, stabilizes or continues to widen, while also building experience with prolonged exposure. It does mean that the 48-month difference should not be interpreted with the same causal certainty as the original randomized result.

The reported five-year finding adds another observation along this trajectory. It does not yet settle the optimal length of treatment, because the extension did not randomly assign people to continue, stop or change treatment after amyloid was reduced.

Safety over time remains part of the benefit calculation

Longer follow-up is particularly valuable for evaluating amyloid-related imaging abnormalities, or ARIA, the characteristic safety concern associated with anti-amyloid antibodies. ARIA is detected by brain imaging and can include swelling or small areas of bleeding.

In the 48-month report, investigators said no new safety signals had emerged. They also reported that ARIA rates were low after the first six months and similar to placebo rates from the controlled period. That pattern suggests the risk is concentrated earlier in treatment rather than accumulating steadily, although the later observations came from the open-label extension rather than a continuing placebo-controlled comparison.

“No new safety signals” does not mean an absence of treatment burden or known risk. Lecanemab was administered intravenously every two weeks in CLARITY-AD, and its use requires clinical selection and imaging surveillance. The extension contributes evidence about repeated exposure, but longer-term safety continues to develop alongside the efficacy record.

What durability means for clinical decisions

For patients and families considering an approved disease-modifying treatment, the practical question is not simply whether lecanemab works at 18 months. It is whether a modest slowing of decline persists long enough to matter against the demands and risks of ongoing therapy.

The peer-reviewed 48-month findings provide evidence consistent with sustained benefit in adults who began treatment during early symptomatic Alzheimer disease. The five-year conference update suggests that this pattern may continue, but it remains a preliminary extension of that published record rather than a new randomized demonstration.

Taken together, the results strengthen the case for discussing lecanemab as a long-term treatment rather than a fixed short course. They also leave treatment duration as an active clinical and research question. The clearest finding so far is not that progression stops, but that the measured gap between treated participants and an estimated untreated course may continue to grow over several years.