The FDA-approved anti-amyloid-β monoclonal antibodies for the treatment of Alzheimer’s disease: a systematic review and meta-analysis of randomized controlled trials

The latest data suggest the prevalence of Alzheimer’s disease (AD) will double in Europe and triple worldwide by 2050 [1]. It becomes a public health predicament in the world, and there is a significant impact on the direct cost of AD to the society.

Previously, the National Institute on Aging and Alzheimer’s Association classified the biomarkers of AD into A (amyloid), T (phosphorylated tau), and N (neurodegeneration): the ATN framework [2]. In other words, the main pathological change in AD is the accumulation of amyloid beta (Aβ) material in the brain, which can occur decades before the onset of clinical symptoms [1‐3]. It may also induce downstream lesions, such as tau phosphorylation and aggregation, leading to neuronal death in AD [2, 4‐8]. In addition, stages of AD can range from cognitively normal to mild cognitive impairment and dementia, which spans a period of years and emphasizes the continuity of the disease [1]. Therefore, it is important to diagnose and treat the disease early to slow down the disease process.

Currently, AD can be treated with non-pharmacologic therapy and pharmacologic therapy. Non-pharmacologic therapy consists of lifestyle changes, and multidomain interventions to prevent cognitive decline [6, 7, 9‐11]. Pharmacotherapy is focused on disease-modifying treatments, including drugs targeting Aβ and Tau proteins, and other target classes such as proteostasis/ protein opathies, epigenetic regulators, synaptic plasticity and neuroprotection, inflammation and infection, metabolism and bioenergetics, vascular and growth factors are also of interest [1]. Among them, monoclonal antibodies (mabs) against tau proteins are aimed at binding to extracellular tau proteins, slowing or preventing their diffusion between cells and thus inhibiting tau protein aggregation and neurofibrillary tangle formation [12]. Phase II trials NCT02871921 and NCT03352557 were conducted to test the efficacy and safety of the semorinemab and gosuranemab. Whereas Aβ is the most common target in phase II and phase III drug development programs, only a few anti-amyloid-β (anti-Aβ) drugs have shown statistically significant cognitive benefits in AD clinical trials, despite a large body of evidence supporting the toxic effects of amyloid [13]. The anti-Aβ agents currently in clinical trials include: aducanumab, lecanemab, gantenerumab, donanemab, β-site Aβ precursor protein cleaving enzyme-1(BACE1), and BACE2, with NCT01760005, NCT03444870, NCT03443973, NCT05533411 all underway. Of all anti-Aβ approaches, passive immunotherapy using anti-Aβ mabs against Aβ has been best tolerated and given its mechanistic selectivity, it has been widely considered as the therapeutic candidate of choice [14]. These anti-Aβ mabs are also associated with downstream effects on tau pathology and neurodegeneration [15]. Among them, the FDA approved only two anti-Aβ mabs, aducanumab and lecanemab. Prior to this, only five drugs were approved by the FDA for clinical treatment, including acetylcholinesterase inhibitors and non-competitive N-methyl-d-aspartic acid (NMDA) receptor antagonists. However, these drugs are unable to alter AD progression, only for partial symptomatic relief [16]. Anti-Aβ drugs can slow the progression of the disease, probably because Aβ is more upstream in the overall pathological process, facilitating early treatment [15, 17].Although there have been several previous analyses of the safety and efficacy of anti-Aβ mabs for the treatment of AD, there have been no separate analyses of FDA-approved monoclonal antibodies. Critically, we included the recently reported lecanemab phase III results [18], which was the basis for the FDA’s accelerated approval. It is the second FDA approved anti-Aβ mabs for AD [19] and may have contributed to showing some statistically significant effects. Therefore, to provide evidence for clinicians, we pooled data from previous RCTs and conducted a meta-analysis to investigate the efficacy and safety of different FDA-approved anti-Aβ mabs for the treatment of AD.

FDA-approved lecanemab and aducanumab are anti-Aβ mabs that can slow the disease process of AD [18], targeting the pathophysiological mechanisms of AD. This is the first meta-analysis of the efficacy and safety of only these two FDA-approved drugs. We found statistically significant improvements in clinical outcomes (CDR-SB, ADCS-ADL-MCI, ADCOMS, ADAS-Cog), neuroimaging (amyloid PET SUVr), and biomarkers (CSF Aβ1-42, CSF P-Tau, CSF T-Tau, plasma A β42/40 ratio, plasma p-tau181) with lecanemab. There was no statistically significant difference in CDR-SB for aducanumab compared with placebo. Conversely, aducanumab contributed to the ADCS-ADL-MCI, ADAS-Cog, neuroimaging, and biomarkers outcomes improvements, except for the absence of accessible data for ADCOMS and plasma Aβ42/40 ratio. Both drugs had elevated adverse effects compared to placebo, which means they were more aggressive.

Prior to 2003, the FDA approved only five drugs for the treatment of AD: tacrine, donepezil, rivastigmine, galantamine and memantine. The first four are acetylcholinesterase (AChE) inhibitors, and memantine is an N-methyl-d-aspartic acid (NMDA) receptor-holding agent. All of these drugs only relieve symptoms and do not slow disease progression. In June 2021, the FDA announced accelerated approval of aducanumab, the first drug approved to slow the progression of AD, and another new FDA approval for AD in nearly 20 years. The first drug used to slow the progression of AD [18, 24]. Aducanumab is a human mab that selectively targets aggregated forms of Aβ, including soluble oligomers and insoluble fibrils [17]. Despite the FDA approval, the effectiveness of aducanumab remains controversial. A phase III clinical trial by Budd et al. [22] was used to test the efficacy of aducanumab. These included two large trials, ENGAGE with 1653 patients and EMERGE with 1643 patients, but trials were terminated early due to the outcome of a futility analysis. One reason for discontinuing the trials was that the primary endpoint (CDR-SB) in ENGAGE was not met. However, no evidence has shown that the early termination of the studies affected the integrity or validity of the results or conclusions from either study. The robustness of the study results was demonstrated by sensitivity and supplementary analyses [22]. In fact, the final data from these two studies showed a greater magnitude of treatment effect compared to the invalid interim data. It is noteworthy that aducanumab caused a large reduction in brain Aβ at the cost of a higher ARIA compared to lecanemab. The study by Jeong et al. also reported a higher incidence of adverse events with aducanumab compared to other mabs. The reason for this may be attributed to different biological mechanisms by which different types of mabs target Aβ, as well as their different selectivity for antibody solubility [25]. Aducanumab partially targets oligomers, while primarily clearing insoluble amyloid plaque, which is associated with vasogenic brain edema, raising the risk of adverse effects.

Subsequent to the FDA’s recent approval of lecanemab in January 2023, supported by a clinical research published in February 2023 [19], we performed this meta-analysis and found for the first time that lecanemab may have better efficacy than aducanumab. Possible reason for the great extent of ameliorative effect may be that lecanemab is a humanized IgG1 anti-Aβ mabs and can selectively bind to large, soluble Aβ protofibrils that are the most neurotoxic and contribute to the pathogenesis of AD [26]. The trial to speed up lecanemab approval was a multicenter, double-blind, phase III trial, with the primary endpoint of CDR-SB at 18 months. At 18 months, the primary regression indicator CDR-SB changed less from baseline to the end of follow-up in the lecanemab group compared to the placebo group, while the remaining indicators (amyloid, tau protein, neurodegenerative lesions) decreased more [18]. Compared to aducanumab, lecanemab had a lower risk of side effect, possibly reason was that it selectively targets the soluble conformation of Aβ (i.e., does not bind to plaque) [13, 27]. According to our study, all clinical outcomes were mildly improved. Similar to our findings, a previous review concluded that mabs statistically improved cognition with small effect sizes and vigorously reduced brain amyloid burden, but increased the risk of ARIA [8]. However, this review lacked the data analysis of lecanemab.

As for neuroimaging, PET SUVr is the only imaging data available for the assessment of Aβ deposition by PET. Previous studies have shown that assessing enrichment of Aβ plaque load is particularly relevant in assessing the feasibility of clinical trials in enriched amyloid-positive patients with AD, where separate clinical criteria appear to lead to serious misclassification [28]. This is in line with the current trend of AD diagnosis and treatment. In the context of the imaging boom, PET-CT can help increase the possibility of early diagnosis of AD and help patients receive treatment before symptoms appear for a better quality of life. In addition, CSF (Aβ1-42, T-Tau, P-Tau) and plasma (p-tau181, Aβ42/40 ratio) from selected patients were collected and analyzed together, and it was found that changes in biomarkers may be sequential in AD patients [22]. Previous studies have shown that an increase in Aβ plaques occurs first, followed by an increase in soluble p-tau levels, which in turn may lead to the accumulation of neurofibrillary tangles (NFTs) and subsequent cognitive decline [29]. Therefore, targeting the upstream of AD pathogenesis for the earlier efficacy to slow down the disease process.

We also have some limitations. Most notably, the number of RCTs we included was small and sample size varied differently. In addition, we only analyzed data from the experimental group at a single dose (10 mg/kg) and failed to take into account the effects of different doses on outcomes, which may reduce the credibility of the results. We chose this single dose (10 mg/kg) because it was the only dose that all of the RCTs included, and it has been identified as an appropriate dose [17]. Moreover, in the most recent and largest RCT, only a biweekly 10 mg/kg dose of lecanemab was used to treat early AD [18]. We performed subgroup analyses of the different outcome indicators according to the therapeutic agents of the included patients. However, subgroup analyses were not performed according to different populations (e.g., women, APOE e4 homozygous carriers), in which the effects may be different than in the whole sample (see, for example, the supplementary material of the van Dyck et al. lecanemab phase III RCT. Another limitation is that the effect of aducanumab on structural MRI (greater ventricular enlargement compared with placebo) was not considered in this review. Greater atrophy induced by these drugs is a potential concern.

Although the FDA approved two drugs to slow the disease process, the safety of these two drugs is yet to be considered and more clinical trials are expected to prove this.

This meta-analysis showed that FDA-approved anti-Aβ mabs statistically improved clinical outcomes and neuroimaging, and statistically changed the levels of biomarkers, suggesting a role for both drugs in slowing disease progression in AD patients, but at the cost of an increased probability of side effects. From this meta-analysis, we found for the first time that lecanemab may have better efficacy than aducanumab. These results offer new hope for the development of anti-Aβ mabs. We also hope that these results will provide a reference for the discovery of targeting the pathological mechanisms of AD, with the aim of developing more effective drugs that can modify the disease process of AD.