Dementia, the disease that erases memories. Among them, Alzheimer's disease, which accounts for more than 70%, long had no adequate treatment. Only recently has a path opened with the advent of antibody therapies that directly remove amyloid beta, the causative substance in the brain.
Representative examples include Leqembi (ingredient lecanemab), developed by Japan's Eisai and U.S. Biogen and approved by the U.S. Food and Drug Administration (FDA) in 2023, and Kisunla (ingredient donanemab), developed by Eli Lilly and Company and approved by the FDA in 2024.
But a fresh debate has arisen recently over anti-amyloid Alzheimer's therapies. It was sparked by the publication of an analysis that reviewed and evaluated past clinical studies of failed anti-amyloid drugs together with clinical studies of the latest drugs.
The study concluded that anti-amyloid agents had minimal therapeutic effect and increased the risk of side effects. Citing these findings, some commentaries questioned not only the effectiveness of the latest treatments but the treatment hypothesis itself.
In particular, in Korea, inquiries from patients have surged at medical sites, and some corporations have moved to use the issue as an investment catalyst.
In response, dementia experts at home and abroad expressed concern and issued a direct rebuttal.
ChosunBiz met with Choi Ho-jin, a neurology professor at Hanyang University Guri Hospital (planning director at the Korean Dementia Association), at a shared office on Teheran-ro in Gangnam-gu, Seoul, on the 28th to examine the controversy.
Professor Choi noted, "The issue this time is not with the drugs themselves but with the study's design and how the data were interpreted." Choi said, "The biggest concern is that provocative and incorrect interpretations are spreading repeatedly, causing patients to miss the window for treatment."
◇ "Limits of bundling first-generation failed drugs with the latest drugs"
The controversy was triggered by a meta-analysis published in the April issue of the international journal issued by Cochrane, known for evidence-based medicine research.
A joint team from Italy, Switzerland, and the Netherlands reported in the Cochrane Database of Systematic Reviews (CDSR) the results of pooling data from 17 clinical trials, involving about 20,000 participants, on anti-amyloid therapies including aducanumab, lecanemab, and donanemab.
The main finding was that anti-amyloid therapies had very small or no effects on improving cognitive function, while the risk of ARIA (brain edema and microhemorrhage), a representative side effect, increased.
Experts at home and abroad immediately voiced concern. On the day the Cochrane review was released, global research institutions including the UK Dementia Research Institute (UK DRI) issued a joint statement saying, "A simple meta-analysis that lumps together different antibodies may distort or underestimate the effects of the latest approved therapies," urging "caution in interpretation."
Scholars at home and abroad pointed out an "average trap," in which the performance of the latest therapies that do have real effects appears statistically diminished because the average includes past drugs that failed in development.
Choi said, "The Cochrane review analyzed 17 studies on anti-amyloid antibody drugs conducted over the past decade or so. Of these, 15 were failed drugs, and only two recent drugs showed meaningful results. Nonetheless, bundling different generations of drugs and averaging them dilutes the effects of the latest therapies."
He added, "It is not scientifically sound to conclude with this kind of design that the latest therapies have no effect."
In fact, early antibody therapies had limited and imperfect amyloid-clearing ability, patient selection, and clinical design. In contrast, Leqembi selected amyloid-positive patients using PET scans and applied high-dose administration and precise designs, partially demonstrating actual amyloid reduction and changes in clinical measures.
◇ "Anti-amyloid new drugs slow dementia progression… ARIA is a manageable risk"
Experts at home and abroad say it is more important to assess effectiveness and safety using real-world data from currently approved therapies.
According to the Alzheimer's Association's 2026 annual report, therapies such as Leqembi (lecanemab) were evaluated as "disease-modifying therapy," which slows the disease process itself by removing amyloid beta from the brain.
The analysis estimated that for patients with mild cognitive impairment (MCI), the average time to progress to dementia was 7.2 years without treatment, but 9.7 years in the treatment group, a delay of about 2.5 years. It also suggested that starting treatment in the early stage could delay progression by up to 13.2 years.
Choi said, "Four-year follow-ups show a tendency for the treatment group to remain longer in the early stage," adding, "The core of dementia treatment is how much we can delay it, and extending the period during which patients can live independently is highly meaningful."
Regarding ARIA, the most concerning side effect, Choi explained that "domestic data are stable." According to the Korean Dementia Association's JOY-ALZ analysis (as of Dec. 2025), 85.2% of 775 patients across 42 institutions received lecanemab. The analysis found an ARIA incidence of 14.8%, with symptomatic ARIA at only 0.6%.
Choi said, "Most adverse reactions—4.1% edema (ARIA-E) and 13.2% hemorrhage (ARIA-H)—were asymptomatic or within a range manageable under clinician monitoring," adding, "There is evidence that anti-amyloid therapy has already moved beyond the research stage and taken root as a standard system in clinical practice." Choi said, "In Korea's medical setting, strict patient selection and monitoring are ensuring safety."
Choi pointed out, "The problem with this controversy is less the study itself than the statements by some experts who offered definitive interpretations—as if anti-amyloid therapy were meaningless—without fully explaining the study's limits."
Choi said, "Even before, pre-treatment explanations took more than an hour; after this controversy, they have become even longer, and clinicians must address patients' and caregivers' anxiety and misunderstandings one by one."
Choi added, "Hasty generalizations and declarative criticisms, spread by some online and offline outside the bounds of academic caution, are adversely affecting real patient care, so we cannot take this issue lightly."
Choi also had a message for early-stage patients and their caregivers.
Choi emphasized, "The most worrisome thing is giving up on treatment or missing the timing based only on controversial information," adding, "It is most important to consult thoroughly with clinicians and begin treatment at the appropriate time."
☞Leqembi (ingredient lecanemab)
This therapy is not for all dementia patients. It is currently used for patients with mild cognitive impairment (MCI) or early Alzheimer's disease with confirmed amyloid beta. Before treatment, PET or cerebrospinal fluid testing is required to confirm amyloid accumulation, and regular MRI monitoring is also necessary.
☞ARIA (Amyloid-Related Imaging Abnormalities)
A collective term for imaging abnormalities observed on MRI associated with anti-amyloid antibody therapy. It is classified into edema/exudation (ARIA-E), which appears as accumulation of fluid in the brain parenchyma or around blood vessels, and hemorrhagic changes (ARIA-H), including microhemorrhages and superficial siderosis.
References
DND (2024), DOI: https://doi.org/10.12779/dnd.2024.23.4.165
CDSR (2026), DOI: https://doi.org/10.1002/14651858.CD016297
The Journal of Prevention of Alzheimer's Disease (2026), DOI: https://doi.org/10.1016/j.tjpad.2026.100562
Alzheimer's Association, 2026 Alzheimer's Disease Facts and Figures