Alzheimer’s disease (AD), the most common form of dementia, is characterized by progressive memory loss, cognitive dysfunction, language impairment, personality and behavioral changes, and declining orientation and judgment. In the final stages, patients lose all ability to care for themselves (Scheltens et al., 2021). Due to its high incidence and significant social impact, AD has become a major global public health challenge. The World Health Organization (WHO) predicts that the number of AD patients worldwide will reach 82 million by 2030 and 152 million by 2050 (Zhang et al., 2020). In China, there are currently six million AD patients, with the number expected to exceed 20 million by 2050. This alarming trend has prompted governments and researchers worldwide to accelerate related studies (Zhang et al., 2020).

The complex clinical manifestations of AD are multidimensional: in addition to core memory impairment, patients often experience aphasia (loss of language function), apraxia (impaired motor skills), agnosia (impaired object recognition), visuospatial deficits, executive dysfunction (impaired planning and decision-making), and significant personality and behavioral changes, ultimately leading to irreversible dementia or death (Weller and Budson, 2018). The pathological mechanisms are based on two hallmark changes: the deposition of neurotoxic amyloid-β (Aβ) plaques in the brain and neurofibrillary tangles (NFTs) caused by hyperphosphorylated tau protein, both of which contribute to neuronal apoptosis and synaptic dysfunction (Mantzavinos and Alexiou, 2017).

Currently, the main pharmacological treatments for AD (Knapskog et al., 2021; Briggs et al., 2016) are cholinesterase inhibitors (ChEIs), including donepezil, rivastigmine, and galantamine, which increase acetylcholine levels in the synaptic cleft and are used to treat mild to moderate AD. These drugs can improve cognitive function, overall impression, daily living abilities, and psychiatric symptoms. Another drug, memantine, an NMDA receptor antagonist, is approved for moderate to severe dementia and can help manage delusions and agitation in patients with moderate to severe AD (Briggs et al., 2016). Other drugs, such as huperzine A and oxiracetam, are also used in AD treatment (Briggs et al., 2016). However, existing drugs only partially alleviate symptoms or slow disease progression, and there is no treatment that can reverse the pathological process.

In recent years, despite the potential shown by Aβ and tau-targeted therapies in animal studies, several international Phase III clinical trials (e.g., Aducanumab, Semorinemab) have failed to meet their endpoints, highlighting the complexity of AD drug development and the challenges of clinical translation (Rostagno, 2022). In this context, disease-modifying therapies (DMT) targeting Aβ, such as monoclonal antibodies, have emerged as a new direction (Narang et al., 2020). Researchers believe that Aβ metabolic imbalance and its neurotoxicity are central to AD pathology, with abnormal aggregation directly damaging neurons and causing cognitive impairment (Ono, 2018; Shi et al., 2022). In January 2023, the anti-Aβ protofibril antibody lecanemab was approved by the FDA for the treatment of early AD, followed by donanemab in July 2024, which significantly delayed cognitive decline and disease progression (Cummings et al., 2023; Kurkinen, 2023). Although both drugs have shown positive results in clinical trials, the association between amyloid-related imaging abnormalities and the APOE ε4 genotype remains controversial. This study aims to integrate all Phase III trial data post-2021 to re-evaluate the clinical efficacy and safety of lecanemab/donanemab, providing more evidence-based decision-making for AD treatment.

Alzheimer’s disease (AD) is a neurodegenerative disorder characterized by progressive cognitive dysfunction and is the leading cause of neurocognitive disorders (NCD) in the elderly (Zheng and Wang, 2025). Its typical clinical manifestations include progressive memory decline (with anterograde and retrograde amnesia being prominent), executive dysfunction, and loss of daily living abilities. The incidence and prevalence of AD increase exponentially with age. The hallmark neuropathological features of AD include extracellular amyloid-β (Aβ) plaques and intracellular neurofibrillary tangles caused by hyperphosphorylated tau protein. Research has confirmed that the Aβ cascade plays a central role in AD pathogenesis, with abnormal aggregation triggering neuronal toxicity, synaptic dysfunction, and neuroinflammation (Delport and Hewer, 2022).

Current clinical management of AD primarily involves multimodal intervention strategies. Traditional pharmacological treatments include two main classes: cholinesterase inhibitors (e.g., donepezil), which enhance cholinergic neurotransmission to improve symptoms, and NMDA receptor antagonists (e.g., memantine), which modulate the glutamatergic system to provide neuroprotection. However, these drugs only temporarily alleviate symptoms and do not slow disease progression. In terms of adjunctive therapies, acupuncture (including conventional needling and electroacupuncture at specific points) and massage techniques are often used to improve cognitive function, potentially by modulating cerebral blood flow and neuroplasticity (Zhang and Chuxiao, 2022).

With advances in understanding AD pathology, disease-modifying therapies (DMT) targeting Aβ have become a focus of research. Current evidence suggests that the imbalance between Aβ production and clearance, leading to abnormal brain deposition, is a key initiating factor in the AD pathological cascade. Aβ oligomers can induce synaptic toxicity, mitochondrial dysfunction, and blood-brain barrier damage, ultimately leading to neuronal loss and cognitive impairment (Walsh and Teplow, 2012; Regland and Gottfries, 1992; Nakano et al., 2022). In this context, the new anti-Aβ monoclonal antibodies Lecanemab and Donanemab have shown significant therapeutic potential.

Lecanemab, a humanized IgG1 monoclonal antibody, exerts its therapeutic effects by specifically binding to soluble Aβ oligomers and protofibrils. Animal studies have shown that Lecanemab significantly reduces the number of pathogenic Aβ plaques in AD models, inhibits Aβ aggregation, and selectively clears Aβ protofibrils from brain tissue and cerebrospinal fluid (Shi et al., 2022). Population pharmacokinetic studies indicate that the drug follows a linear two-compartment model: compared to a 10 mg/kg monthly dosing regimen, biweekly dosing more rapidly reduces amyloid PET standardized uptake value ratios (SUVr) and plasma p-tau181 levels, while significantly increasing the Aβ42/40 ratio. Notably, the half-life of amyloid re-accumulation in the brain after treatment cessation is up to 4 years, while plasma biomarkers (Aβ42/40 and p-tau181) recover much faster than brain pathology (Hayato et al., 2022).

Donanemab, a recombinant humanized IgG1 monoclonal antibody developed by Eli Lilly, works by specifically recognizing and clearing deposited Aβ plaques. The drug was approved by the FDA in July 2024 for the treatment of early AD (mild NCD or mild dementia stages), making it the first Aβ-targeted therapy that does not require lifelong administration. Its innovative treatment strategy involves periodic dosing to achieve Aβ plaque clearance, and it has not yet entered the Chinese market (Kang, 2024).

This study systematically evaluated the clinical efficacy and biological effects of anti-Aβ monoclonal antibodies (Lecanemab/Donanemab) on patients with early AD through a meta-analysis. The results showed that the intervention group had significant advantages in core indicators such as the Alzheimer’s Disease Composite Score (ADCOMS), Clinical Dementia Rating-Sum of Boxes (CDR-SB), Alzheimer’s Disease Assessment Scale-Cognitive Subscale (ADAS-Cog 14), and amyloid PET load, indicating that these drugs have dual effects of improving cognitive function and clearing pathological proteins. In terms of ADCOMS score, the intervention group had a reduction of 0.05 units compared with the control group (95% CI: 0.07 to −0.03), which is consistent with the disease-modifying effect observed in the AHEAD 3–45 study (Raman et al., 2024). ADCOMS, as a composite endpoint integrating cognition, function, and biomarkers, may be improved due to the indirect regulatory effect of Aβ clearance on synaptic plasticity (DeVos et al., 2017). The CDR-SB score decreased by 0.49 units (95% CI: 0.67 to −0.30), especially in the dimension of self-care ability, suggesting that anti-Aβ treatment may delay the degenerative changes in functional brain regions (such as the default mode network) by reducing Aβ-mediated neuronal network damage (Singh et al., 2023). It is worth noting that the cognitive improvement amplitude shown by the ADAS-Cog 14 scale (SMD = −1.06) was significantly higher than that of traditional cholinesterase inhibitors (usually 0.3–0.5), which is in line with the efficacy trend of Lecanemab in the Clarity AD study (van Dyck et al., 2023), indicating that disease-modifying treatment may break through the therapeutic bottleneck of symptomatic treatment.

PET imaging analysis showed that the amyloid load in the intervention group decreased by 72.99 SUVr units (95% CI: 88.58 to −57.41), which may be related to the high-affinity clearance characteristics of Donanemab for mature plaques. However, there is a dissociation between amyloid clearance and the magnitude of cognitive improvement (e.g., only a 1.06 unit improvement in ADAS-Cog), a phenomenon also reported in the TRAILBLAZER-ALZ four trial (Mintun et al., 2021). This may reflect the following mechanisms: (1) Aβ clearance requires multiple steps, such as synaptic remodeling and neuroinflammatory relief, to be translated into clinical benefits; (2) The continuous progression of tau pathology may offset the potential benefits of Aβ clearance; (3) The existing scales are not sensitive enough to capture the subtle changes in early AD.

This study focused on the risk of amyloid-related imaging abnormalities (ARIA) and other adverse events. The results showed that anti-Aβ treatment significantly increased the risk of ARIA, with specific associations between its subtypes and genetic susceptibility, which has important guiding significance for clinical risk stratification and drug monitoring.

The occurrence of ARIA-E (edema/effusion) and ARIA-H (hemosiderin deposition) is closely related to the mechanism of drug clearance of intracerebral Aβ. Lecanemab targets soluble Aβ protofibrils, while Donanemab clears dense plaques. Both may accelerate the stripping of Aβ around blood vessels, leading to damage to the vascular basement membrane structure. This study found that anti-Aβ treatment increased the overall risk of ARIA by 4.35 times (RR = 4.35), with the most significant increase in the risk of ARIA-E (RR = 8.78). It is worth noting that the APOE ε4 genotype has a bidirectional regulatory effect on the risk of ARIA-E: non-carriers have a 10.97-fold increase in risk, while homozygous patients have a 10.84-fold increase in risk. This suggests that APOE ε4 may affect the occurrence of ARIA through a dual mechanism—both by exacerbating the severity of cerebral amyloid angiopathy (CAA) and by amplifying the blood-brain barrier disruption effect through apolipoprotein E-mediated inflammatory responses. Although the risk of ARIA-H (RR = 2.21) is lower than that of ARIA-E, the subtype analysis shows that the risk of superficial siderosis (RR = 2.74) is significantly higher than that of microbleeds (RR = 1.97). This may be related to chronic vascular leakage induced by drugs: local inflammatory responses (such as complement activation and cytokine release) triggered during Aβ clearance can disrupt the integrity of the blood-brain barrier, leading to red blood cell extravasation. After chronic leakage, iron ions produced from the breakdown of hemoglobin exceed the brain’s clearance capacity (such as phagocytosis by glial cells or cerebrospinal fluid drainage), eventually depositing in the leptomeninges and brain surface. This mechanism is supported by the 2.63-fold increase in the risk of central nervous system superficial siderosis in this study (RR = 2.63, P < 0.0001). Although anti-Aβ treatment did not significantly increase the risk of conventional adverse events (such as falls and dizziness) (RR = 1.12, P = 0.02), the ARIA-related risk needs to be managed through the following strategies: adjusting monitoring frequency based on APOE ε4.

APOE genotype (particularly the ε4 homozygous status) may play a pivotal role in balancing the efficacy and safety of anti-Aβ monoclonal antibodies. Our study revealed that Alzheimer’s disease patients carrying the APOE ε4 homozygous allele exhibited significantly higher ARIA-E incidence following lecanemab/donanemab treatment compared to non-carriers, with a relatively attenuated risk increase observed in the lecanemab group. This phenomenon may be linked to the dose-dependent effects of the ε4 allele: ε4 homozygotes likely experience accelerated accumulation of monoclonal antibodies in brain parenchyma due to compromised blood-brain barrier integrity, thereby amplifying vascular edema risks triggered by Aβ-targeted clearance. These findings suggest that dissociation between biomarker response and clinical outcomes may be modulated by genotype. Consequently, we propose implementing a stratified management strategy for ε4 homozygous patients in clinical practice: pretreatment genetic screening could serve as a risk stratification tool, favoring gradual dose-escalation regimens (e.g., extended titration protocols for lecanemab) alongside intensified MRI monitoring (e.g., baseline scans followed by monthly evaluations during the first 3 months) to enable early ARIA detection (cite relevant imaging guidelines). However, current conclusions are constrained by the limited sample size of ε4 homozygotes (representing only X% of the included population). Future research should validate genotype-efficacy/safety causality and explore precision dose optimization through cross-trial individual patient data (IPD) meta-analyses.

In addition, aducanumab is also an effective drug for the treatment of AD. By comparison, it is suggested that the core features of Donanemab/Lecanemab, Aducanumab and symptomatic therapy: In terms of mechanism, Donanemab/Lecanemab targets amyloid β-protein (Aβ) fibrils and oligomers (such as TRAILBLAZER-ALZ two test) (Mintun et al., 2021), while Aducanumab mainly removes deposited plaques (EMERGE/ENGAGE subgroup analysis). Symptomatic therapy (such as acetylcholinesterase inhibitors) only regulates neurotransmitters; in terms of efficacy and safety, Donanemab showed that iADRS score delayed 35% cognitive decline in phase III trials (low/medium Tau subgroup was better) (Mintun et al., 2021), and Lecanemab (Clarity AD test) reduced CDR-SB by 27% (van Dyck et al., 2023), but both were accompanied by ARIA risk (edema 12.6%–17.3%). The incidence of ARIA in Aducanumab is higher (35% edema) and the efficacy is controversial. At the level of clinical applicability, Donanemab/Lecanemab requires strict screening of early AD patients (Aβ + and Tau below medium), high treatment costs (such as Lecanemab annual cost > USD 26,000) and reliance on biomarker monitoring (van Dyck et al., 2023). Although symptomatic therapy is more suitable for a wider population, it cannot delay disease progression. These differences highlight the importance of precise treatment and risk-benefit trade-offs in AD management.

The analysis in this study revealed significant differences in the safety profiles between lecanemab and donanemab, particularly demonstrating a higher ARIA-E risk in the donanemab group. This finding carries direct clinical implications: For patient populations who are APOE ε4 carriers or require rapid amyloid plaque clearance (e.g., rapidly progressive Alzheimer’s disease), clinicians should prioritize evaluating the short-term efficacy-risk balance of donanemab and consider increasing MRI monitoring frequency during pretreatment and early treatment phases (e.g., baseline and scans at months 3, 6, and 9) to dynamically manage ARIA events. These results highlight that drug selection should integrate patient preferences, anticipated treatment duration, and healthcare resource accessibility. For instance, lecanemab’s gradual dose-escalation regimen may be preferable for older adults with heightened sensitivity to tolerability, whereas donanemab might better align with clinical trial objectives requiring rapid biomarker improvements in the short term. Although the current sample size is limited by the predominance of Phase II studies, these findings provide a prospective framework for risk-stratified design in Phase III trials and real-world prescribing protocols. Future studies incorporating large-scale long-term follow-up data could further validate the generalizability of these clinical insights.

In summary, anti-Aβ monoclonal antibodies (Lecanemab/Donanemab) have demonstrated significant efficacy in improving cognitive function (as measured by the ADAS-Cog 14 scale) and activities of daily living (as assessed by the CDR-SB) in patients with early-stage Alzheimer’s disease, and have effectively reduced cerebral amyloid deposition (confirmed by PET imaging), thereby establishing their clinical value as disease-modifying treatments. However, the risk of amyloid-related imaging abnormalities (ARIA) associated with these treatments warrants heightened vigilance, particularly the elevated risk of ARIA-E in APOE ε4 carriers, as well as the occurrence of vascular complications such as superficial siderosis. Based on the current evidence, it is recommended that in clinical practice, APOE ε4 genotyping and baseline imaging assessments be integrated to develop individualized dosing and monitoring plans. Future research should focus on expanding the population coverage (especially including Asian cohorts) and extending follow-up periods to clarify long-term safety.