Two years ago, the anti-amyloid-beta (Aβ) disease-modifying therapy (DMT) lecanemab received marketing approval in China, opening a new chapter in disease-targeted treatment of Alzheimer’s disease (AD) in this country [1]. Over the past two years — from clinical trials to real-world practice, from intravenous infusion to the launch of a subcutaneous formulation (the SC formulation was approved by the U.S. Food and Drug Administration on July 14 for the initial treatment of early AD, and was approved for pilot use in Hainan’s Boao Lecheng on August 5 in China) [2,3], and from single-center exploration to multi-center collaborative networks — China’s AD DMT practice has completed a key leap from making the drug available to making it broadly accessible, and from achieving an initial breakthrough to building a sustained, deep implementation framework.
On July 25, 2026, Prof. Lawrence S. Honig of Columbia University visited China and, based on 380 patients treated at his center between May 2023 and July 2026, delivered a lecture titled “ANTI-AMYLOID THERAPY IN CLINICAL PRACTICE: Focus on Lecanemab.” In the lecture, Prof. Honig systematically shared real-world experience on the long-term safety and tolerability of lecanemab, and — focused on treatment discontinuation and drug choice for anti-Aβ therapy — proposed a clinical-pathway framework grounded in mechanistic differences.
3 Years of Real-World Experience Confirm: Lecanemab Has Favorable Safety and Tolerability
AD is a chronic, progressive disease, and Aβ deposition begins decades before clinical symptoms appear [4]. Prof. Honig noted that this time dimension dictates that anti-Aβ therapy should be planned as long-term maintenance, not as a short-term intervention.
In the practice at Columbia University Irving Medical Center, treatment access followed the FDA label, without adding extra restrictive criteria such as "appropriate use." Among the 380 patients, ages ranged from 35 to 90 years (mean 73.9 years); 70% had a Clinical Dementia Rating-Global Score (CDR-global) of 0.5 and 30% a score of 1; the mean Mini-Mental State Examination (MMSE) score was 23.8. For biomarker confirmation, 67% were confirmed by cerebrospinal fluid (CSF) testing and 33% by positron emission tomography (PET), with 16% confirmed by both. For safety monitoring, the center adopted a “group-by-group prescription” strategy: MRI was scheduled after every 1–2 infusions, rather than issuing long-term prescriptions at once.
The safety analysis based on these patients showed that amyloid-related imaging abnormalities — edema/effusion (ARIA-E) and hemorrhage (ARIA-H) — each occurred in approximately 11% of patients; however, most ARIA-E cases were asymptomatic, with only 4 cases being symptomatic. In addition, APOE genotype was clearly associated with ARIA-E risk. Prof. Honig noted that these safety characteristics are consistent with phase 3 clinical trial data.
In the analysis of reasons for discontinuation, only about 10% of patients stopped treatment due to cognitive decline — a figure Prof. Honig explicitly noted is below the natural-course expectation, suggesting that disease progression is effectively slowed in the real world. A further 30 patients paused due to medical issues (including 6 cerebral infarctions and 2 subdural hematomas), and 19 withdrew due to treatment burden, insurance or personal preference.
Clinical Reasoning: Should Anti-Aβ Therapy Involve “Discontinuation”?
Based on three years of real-world clinical practice covering 380 patients, Prof. Honig raised a deep question that runs through the long-term management of DMT: should anti-Aβ therapy involve “discontinuation”? He developed a penetrating theoretical analysis. He pointed out that the answer to whether to discontinue depends on the molecular target of the drug. For lecanemab, the answer is: it should not be stopped.
The target of lecanemab is soluble Aβ protofibrils [5], and Alzheimer’s disease is a slowly progressive disease that lasts more than 20 years, with Aβ protofibrils continuously produced throughout the course [6]; even if plaques are cleared, this accumulation process and neuronal injury do not stop on their own. Prof. Honig pointed out explicitly that after anti-Aβ therapy is discontinued, PET can still detect re-accumulation of Aβ in patients’ brains at a rate of about 3 centiloids per year. Therefore, as a chronic disease whose course can span more than 20 years, the treatment logic for AD should be analogous to that for cardiovascular disease or diabetes — even when symptoms are relieved, long-term maintenance therapy is still required. Lecanemab therefore has “no biological reason to discontinue.”
At the same time, Prof. Honig specifically emphasized that, at the individual-patient level, there is currently insufficient evidence to prove that the faster the PET signal decreases or the higher the degree of plaque clearance, the greater the clinical benefit [7]. In other words, although amyloid clearance is important evidence that the drug is acting on its intended pathological target, it cannot directly replace cognitive and clinical outcome assessment. Some patients may continue to experience cognitive decline even after reaching PET negativity; in others, some residual plaque may still be seen on PET while the clinical state remains relatively stable. Therefore, amyloid PET is suitable as a diagnostic and pharmacodynamic biomarker, but is not a direct indicator of disease activity. A single PET scan or any single centiloid value is not sufficient on its own to decide whether a patient should continue treatment [7]. Combining in vitro and animal data, Prof. Honig further noted that low-molecular-weight Aβ species (such as oligomers and protofibrils) are themselves neurotoxic, while plaques themselves have no direct injury evidence in humans [8,9]. Therefore, continuous use of lecanemab, which targets protofibrils, is very important for the long-term management of AD.
Worth noting is that the lecanemab subcutaneous formulation has now received priority review status for marketing authorization in China. On August 5, 2026, relying on the Hainan Boao Lecheng International Medical Tourism Pilot Zone’s pilot use policy, the lecanemab subcutaneous auto-injector was approved by the Hainan Provincial Drug Administration at the Hainan Hospital of Ruijin Hospital, Shanghai Jiao Tong University, for the treatment of early AD [3,10]. Prof. Honig expressed high expectations for this new dosage form. He added key pharmacokinetic data, noting that at matched doses, subcutaneous and intravenous administration produce equivalent plasma concentrations, and ARIA incidence is not significantly different, suggesting consistent brain exposure. Prof. Honig emphasized that, from intravenous infusion to subcutaneous injection, lecanemab has achieved a restructuring of the long-term management model through this formulation breakthrough — the launch of the subcutaneous formulation will extend long-term maintenance therapy from hospital infusion centers to patients’ home settings, improving treatment convenience and providing a feasible pathway for long-term AD management.
Summary
When ARIA risk is manageable and patient adherence is good become the norm, the core question of anti-Aβ therapy shifts from “can it be used” and “how well does it work” to "should it be used long-term” and “can it be discontinued.” Prof. Honig’s answer, grounded in target biology, is clear: faced with continuously produced Aβ protofibrils and a disease course spanning decades, maintenance therapy with lecanemab is essential. With the improved accessibility of the subcutaneous formulation, this long-term management model is moving from single-center exploration into broader clinical practice, providing a feasible path for the normalized implementation of disease-targeted AD treatment.
References
[1] National Medical Products Administration. NMPA approves marketing of lecanemab injection. https://www.nmpa.gov.cn/zhuanti/cxylqx/cxypxx/20240109192251169.html?type=pc&m=
[2] Eisai China. U.S. FDA approves Leqembi (lecanemab) subcutaneous auto-injector for the initial treatment of early Alzheimer’s disease. https://www.eisai.com.cn
[3] The new AD drug lecanemab (Leqembi) subcutaneous auto-injector approved in Boao Lecheng. https://www.capwhale.com/newsfile/details/20260806/f7c292c77d374997aa91fb09053903f0.shtml
[4] Wang Y, et al. Community-based three-level comprehensive prevention and treatment of Alzheimer’s disease — Chinese expert consensus (2025 edition). Chinese Journal of Geriatrics. 2025;44(3):227–237.
[5] Tang Y, et al. Expert consensus on disease-modifying therapy for Alzheimer’s disease (2025 edition). National Medical Journal of China. 2025;105(19):1492–1502.
[6] Committee of Experts on Health Technology Promotion of the China National Health Association, et al. Expert consensus on early diagnosis and treatment of AD-source cognitive impairment (2025). Chinese Journal of Medical Imaging. 2025;33(4):337–346.
[7] Honig LS. ANTI-AMYLOID THERAPY IN CLINICAL PRACTICE: Focus on Lecanemab. Presented at Shanghai; July 2026.
[8] Perry R, Kipps C, Soto Martín ME, et al. Lecanemab for treatment of individuals with early Alzheimer’s Disease (AD) who are apolipoprotein E ε4 (ApoE ε4) non-carriers or heterozygotes. J Prev Alzheimers Dis. 2026;13(4):100507. doi:10.1016/j.tjpad.2026.100507
[9] Johannesson M, Söderberg L, Zachrisson O, et al. Lecanemab demonstrates highly selective binding to Aβ protofibrils isolated from Alzheimer's disease brains. Mol Cell Neurosci. 2024;130:103949. doi:10.1016/j.mcn.2024.103949
[10] Eisai China. Marketing authorization application for lecanemab subcutaneous formulation granted priority review. https://www.eisai.com.cn