Comparative Efficacy,Tolerability,and Acceptability of Donanemab,Lecanemab,Aducanumab,Melatonin,and Aerobic Exercise for a Short Time on Cognitive Function in Mild Cognitive Impairment and Mild Alzheimer’s Disease: A Systematic Review and Network Meta-Analysis

Abstract

Background:

The Food and Drug Administration (FDA) has approved lecanemab and aducanumab and is reviewing donanemab, but they have questionable efficacy, serious side effects and are costly, whereas melatonin administration and aerobic exercise for a short time may overcome these problems.

Objective:

We aim to compare the efficacy on cognitive function, tolerability and acceptability of melatonin administration and aerobic exercise for a short time with donanemab, lecanemab, and aducanumab in people with mild AD and MCI.

Methods:

We systematically reviewed relevant randomized placebo-controlled trials (RCTs) in PubMed, the Cochrane Library, CINHAL, and ClinicalTrials.gov and performed network meta-analyses.

Results:

The analysis included 10 randomized placebo-controlled trials with 4,599 patients. Although melatonin and aerobic exercise for a short time were significantly more effective than donanemab, lecanemab, aducanumab and placebo in the primary analysis, there was significant heterogeneity. In the sensitivity analysis excluding exercise, melatonin was significantly more effective than donanemab, lecanemab, aducanumab and placebo, with no significant heterogeneity. Aerobic exercise for a short time was significantly less acceptable than donanemab, aducanumab and placebo. Donanemab, lecanemab, and aducanumab were significantly less tolerable than placebo and donanemab and lecanemab were significantly less acceptable than placebo.

CONCLUSIONS:

Melatonin may be a better potential disease-modifying treatment for cognitive decline in mild AD and MCI. Aerobic exercise for a short time might also be better than donanemab, lecanemab and aducanumab if continued, as it is well tolerated and more effective, although less valid due to heterogeneity. Another limitation is the small number of participants.

Keywords

Aducanumab Alzheimer’s disease donanemab exercise melatonin meta-analysis

INTRODUCTION

Dementia affects approximately 52 million people worldwide, with a reported disability-adjusted life year of approximately 25 million years [1]. The development of disease-modifying treatments for Alzheimer’s disease (AD), the most common cause of dementia, is therefore a pressing issue. Until recently, the Food and Drug Administration (FDA)-approved treatments for AD were limited to symptomatic treatments such as cholinesterase inhibitors and the N-Methyl-D-aspartate receptor antagonist. However, based on the amyloid-β (Aβ) hypothesis, which states that the accumulation of Aβ causes abnormal phosphorylation of the tau protein, leading to neurodegeneration and cognitive decline [2], anti-Aβ antibody therapies have been developed. The FDA has (accelerated) approved lecanemab and aducanumab and is reviewing donanemab, which is highly controversial because of their small effect sizes, which raises doubts about their clinical significance [3]. In addition, they cause serious side effects such as amyloid-related imaging abnormalities (ARIA) [4], and a recent meta-analysis showed a significant correlation between the administration of anti-Aβ antibody therapies and accelerated ventricular volume enlargement and brain atrophy [5], suggesting treatment related harm [3]. They are also too expensive, with lecanemab costing $26,500 per year [6] and aducanumab $28,200 per year [7]. In addition, the suspicion of fabrication was raised in an article that provided important evidence for the amyloid hypothesis [8].

On the other hand, several modifiable lifestyle risk factors for AD have been reported [9]. In this study, we focused on melatonin administration and aerobic exercise because they are similar to donanemab, lecanemab, and aducanumab in terms of effective potential disease-modifying treatments, but different in terms of being inexpensive with fewer side-effects for the treatment of cognitive decline in patients with mild AD and mild cognitive impairment (MCI). Among lifestyle changes that may be disease-modifying treatments for mild AD and MCI, melatonin and aerobic exercise have shown more evidence: Exercise has been reported to reduce Aβ accumulation and excessive phosphorylation of tau protein, as well as reduce oxidative stress and suppress neuroinflammation [10, 11]. Oxidative stress has been suggested to promote the accumulation, which in turn leads to microglial activation, resulting in increased neuroinflammation and apoptosis [12]. In addition, Rody et al. [13] suggested that exercise increases the secretion of exerkines such as irisin, lactate, and IL-6, which in turn stimulate the secretion of brain-derived neurotrophic factor (BDNF), resulting in neurogenesis and suppression of neuroinflammation. In addition, previous meta-analyses of randomized placebo-controlled trials (RCTs) of cognitive function in AD found exercise to be significantly superior to control or placebo [14]. A meta-analysis by Lopez-Ortiz et al. [15] compared the effects of aerobic exercise and muscle training separately, mixed training with both, and a control group on cognitive function in patients with AD and found that aerobic exercise alone was significantly better than the control group. Additionally, a meta-analysis by Zhang et al. [14] suggested that 30 minutes or less of aerobic exercise per session was more effective than for a longer periods. Therefore, we focused on aerobic exercise for 30 minutes or less in this study.

Similarly, melatonin, a circadian-regulating hormone from the pineal gland that has been shown to be safe at high doses [16], reduces Aβ accumulation and inhibits abnormal phosphorylation of tau protein in animal models of AD [17 –19], and also has antioxidant, anti-inflammatory and BDNF-secretion-enhancing effects [20]. Cerebrospinal fluid melatonin levels decrease with age, with a marked decrease in AD patients [21]; melatonin 1 and melatonin 2 receptors have also been found to decrease in AD patients [22, 23]. A meta-analysis of RCTs suggested that melatonin was significantly more effective than placebo in improving cognitive function in patients with AD [24].

To our knowledge, no studies have compared the effects of donanemab, lecanemab, aducanumab, melatonin, and short-duration aerobic exercise on cognitive function in patients with mild AD and MCI. We hypothesized that melatonin and short-duration aerobic exercise would have a superior cognitive effect, tolerability, and acceptability than donanemab, lecanemab, and aducanumab in mild AD and MCI and tested this hypothesis using a systematic review and network meta-analysis. The aim of this study was to evaluate alternative disease-modifying therapies that could compensate for the shortcomings of donanemab, lecanemab, and aducanumab, and since the FDA indications for lecanemab and aducanumab are limited to mild AD and MCI, this study focused on mild AD and MCI.

METHODS

The protocol of this systematic review and meta-analysis was pre-registered in PROSPERO (number: CRD42022304807) and followed the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) –network meta-analysis guideline (Supplementary Table 1) [25]. In addition to the protocol, we also assessed the efficacy, tolerability, and acceptability of donanemab and the other drugs, which were included in the NMA.

Search strategy and selection criteria

We searched The Cochrane Library, PubMed, CINAHL, and ClinicalTrials.gov from inception to 25 December, 2023, using the following search formula: (donanemab OR LY3002813 OR lecanemab OR BAN2401 OR aducanumab OR BIIB037 OR “aerobic exercise” OR “aerobic training” OR “cardio exercise” OR melatonin) AND (“Alzheimer’s disease” OR “Alzheimer disease” OR dementia OR MCI OR “mild cognitive impairment”) AND cogni* AND random* for The Cochrane Library, PubMed, CINAHL, while (lecanemab OR BAN2401 OR aducanumab OR BIIB037 OR “aerobic exercise” OR “aerobic training” OR “cardio exercise” OR melatonin) AND (“Alzheimer’s disease” OR “Alzheimer disease” OR dementia OR MCI OR “mild cognitive impairment”), with the status of “with results,” for ClinicalTrials.gov. We defined the following inclusion criteria: 1) randomized placebo-controlled trials assessing the efficacy of melatonin, short-duration aerobic exercise, lecanemab, and aducanumab on cognitive function in people with mild cognitive impairment and mild Alzheimer’s disease, and 2) studies setting cognitive function assessment as the primary outcome. However, because melatonin is a sleep medication, it was expected that cognitive function would rarely be assessed as a primary outcome; therefore, we included studies that assessed cognitive function as a secondary outcome for melatonin as an exception. 3) Studies in which participants had a mean or median Mini-Mental State Examination (MMSE) score of at least 18 points or a Clinical Dementia Rating (CDR) global score of less than 1 point. 4) For aducanumab, a high-dose group was included, and for lecanemab, an intervention of 10 mg per kg body weight once every 2 weeks was included. Previous studies have reported that only the above arms were significantly superior to placebo for cognitive function in MCI and mild AD [26, 27], which is why they were approved by the FDA. Similarly, for exercise, we included studies with a single intervention duration of 30 min or less, as described in the introduction. In other words, the evaluation was conducted on the same ground, i.e., using the methods that were reported to be most effective for each intervention. 5) We also included studies where the exercise was clearly described as aerobic in the paper. 6) We included studies in which patients were already taking symptomatic medications for AD (e.g., donepezil or memantine), but sufficient time had elapsed before the start of the trials. 7) Only studies published in English were included in this analysis. The following exclusion criteria were applied: 1) Studies that focused on a group of patients with specific medical conditions, such as glucose intolerance. 2) Studies that focused on specific cognitive assessments, such as spatial or verbal memory, but did not assess overall cognitive function. 3) Studies that used concomitant psychological interventions, such as cognitive training. 4) Studies in which drug or exercise interventions were part of a combined intervention. For example, a drug combination of melatonin and tryptophan, or a multicomponent exercise intervention that included aerobic exercise. 5) Studies that focused on MCI caused by diseases other than AD, such as vascular MCI, were excluded. 6) Studies that did not describe or could not calculate the data required for the network meta-analysis (number of participants and mean or standard error of change in cognitive function before and after the intervention). One reviewer (I.T.) applied the eligibility criteria and selected studies for inclusion in the systematic review, while another reviewer (W.K.) checked the decision. Disagreements were resolved by discussions.

Data extraction and quality assessment

The name of the first author, year of publication, diagnostic criteria, number of patients, age, duration of the study, rating scale used in the study, value of change in cognitive function before and after the intervention (continuous variables), drop-out rates due to adverse events and drop-out rates due to any reason were extracted by one reviewer (I.T.), while another reviewer (W.K.) checked the data received. If data were missing, the original author was not contacted. Data were stored in Microsoft Excel. I.T. assessed the risk of bias using RoB 2.0 [28] and the certainty of evidence using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) framework [29], and W.K. checked the decisions. Disagreements were resolved by discussion. Publication bias was assessed when the number of arms for each comparison exceeded 10 [30].

Outcomes

The primary outcomes were comparison of changes in cognitive function before and after the intervention, relative tolerability (drop-out rates due to adverse events) and relative acceptability (drop-out rates due to any reason). As in our previous study, the MMSE was used first, followed by the CDR scores, as a measure of cognitive function [31]. Standard mean difference was used for continuous variables and odds ratios for dichotomous variables.

Data analysis

We performed a frequentist random-effects network meta-analysis using the “netmeta” package [32] and R version 4.2.1 software [33]. Network meta-analysis is a statistical method for simultaneously comparing three or more interventions by combining direct and indirect evidence [34]. Integrating direct and indirect evidence can usually improve precision [34]. For evidence from indirect comparisons to be valid, the transitivity assumption needs to be met [34]. Concerns remain about the transitivity assumption in this study because a non-pharmacological treatment, exercise, was compared with a pharmacological treatment, and the results should be interpreted with caution. Because of this conceptual heterogeneity, we conducted a sensitivity analysis that excluded exercise and focused only on pharmacotherapy. Furthermore, total heterogeneity/inconsistency in the network was assessed quantitatively by I² values and also tested for statistical significance. Total heterogeneity/inconsistency was then decomposed into within-design (i.e., heterogeneity) and between-design (i.e., inconsistency) and the statistical significance of each Q-value was assessed. Moreover, global and local inconsistency was assessed using the design-by-treatment test and the node-splitting method [35, 36]. Reasons for statistical heterogeneity or inconsistency were explored by sequentially excluding each intervention comparison with a higher Q value from the analysis until the p-value was no longer significant. If inconsistency/heterogeneity was still not apparent, a leave-one-out analysis was conducted.

RESULTS

Identification of relevant studies

As a result of the systematic review, 885 studies were screened and 10 (11 trials) were finally included (Supplementary Figures 1 and 2) [26 , 37–44]. No studies directly compared donanemab, lecanemab, aducanumab, melatonin, and aerobic exercise for a short time. A total of 6,592 patients were included. Further clinical characteristics are described in Table 1. Due to the small number of included studies, we added a posteriori sensitivity analyses that changed the hierarchy of measurement indicators: among the included studies, MMSE remained at the top of the hierarchy, as it is the most commonly used and some studies used only MMSE as a measurement. Alzheimer’s Disease Composite Score (ADCOM) or Alzheimer’s Disease Assessment Scale-cognitive subscale (ADAS) were then chosen as the next indicator to be used.

Table 1

Clinical characteristics of the included studies

Study	Diagnostic criteria	Intervention	Number of patients	Age (mean or median [SD] or [range])	Baseline cognitive function (mean [SD])	Duration	Measurement scales	Results
Mintun et al. (2021) [39]	N/A	Placebo monthly	126	75.4 [5.4]	23.7 [2.9]	76 weeks	MMSE, CDR-SB, ADAS-Cog
		Donanemab 1400 mg monthly	131	75.0 [5.6]	23.6 [3.1]			Negative
Sims et al. (2023) [40]	N/A	Placebo monthly	876	73.0 [6.2]	22.2 [3.9]	76 weeks	MMSE, CDR-SB, ADAS-Cog
		Donanemab 1400 mg monthly	860	73.0 [6.2]	22.4 [3.8]			Positive
Swanson et al. (2021) [41]	NIA-AA core clinical criteria	Placebo biweekly	238	72.0 [50–89]	2.9 [1.5]	18 months	CDR-SB, ADCOMS, ADAS-Cog
		Lecanemab 10 mg/kg biweekly	152	73.0 [51–88]	3.0 [1.4]			Negative
van Dyck et al. (2023) [27]	NIA-AA core clinical criteria	Placebo biweekly	875	71.0 [7.8]	3.2 [1.3]	18 months	CDR-SB, ADCOMS, ADAS-Cog
		Lecanemab 10 mg/kg biweekly	859	71.4 [7.9]	3.2 [1.3]			Positive
Budd Haeberlein et al. (2022) [26]	NIA-AA core clinical criteria	Placebo monthly	548	70.8 [7.4]	26.4 [1.8]	78 weeks	MMSE, CDR-SB, ADAS-Cog
		Aducanumab 10 mg/kg monthly	547	70.6 [7.5]	26.3 [1.7]			Positive
	NIA-AA core clinical criteria	Placebo monthly	545	69.8 [7.7]	26.4 [1.7]	78 weeks	MMSE, CDR-SB, ADAS-Cog
		Aducanumab 10 mg/kg monthly	555	70.0 [7.7]	26.4 [1.8]			Negative
Wade et al. (2014) [43]	N/A	Placebo daily	29	77.3 [6.6]	21.4 [4.7]	6 months	MMSE, ADAS-Cog
		Melatonin 2 mg daily	32	73.5 [8.6]	22.1 [3.5]			Positive
Xu et al. (2020) [44]	DSM-IV	Placebo daily	39	66.5 [8.3]	20.6 [2.2]	6 months	MMSE
		Melatonin 0.15 mg/kg daily	40	66.3 [8.8]	23.2 [2.1]			Positive
Arcoverde et al. (2014) [37]	NINCDS-ADRDA	Control	10	79.0 [74.7–82.2]	19.9 [3.4]	16 weeks	MMSE, CAMCOG
		Aerobic exercise (On a treadmill for 30 min, twice a week)	10	78.5 [64–81.2]	20.4 [2.7]			Positive
Enette et al. (2020) [38]	DSM-IV-TR	Control	21	79 (75–84)	21 [17–23]	9 weeks	MMSE
		Aerobic exercise (On an ergocycle for 30 min, twice a week)	14	74 (68–83)	18 [16–21]			Positive
		Aerobic exercise (On an ergocycle for 30 min, twice a week)	17	79 (75–82)	18 [17–19]			Positive
Varela et al. (2012) [42]	The consensus of the Spanish Society of Geriatrics and Gerontology	Control	15	The mean age of the three groups was 78.3 [9.5].	21.8 [3.2]	12 weeks	MMSE
		Aerobic exercise (On a cycle for 30 min, three times a week)	27		19.3 [6.1]			Negative
		Aerobic exercise (On a cycle for 30 min, three times a week)	26		20.6 [5.4]			Negative

ADAS-Cog, Alzheimer’s Disease Assessment Scal-cognitive subscale; ADCOMS, Alzheimer’s Disease Composite Score; CAMCOG, Cambridge Cognitive Examination; CDR-SB, Clinical Dementia Rating Sum of boxes; DSM-IV, The Diagnostic and Statistical Manual of Mental Disorders the 4th edition; DSM-IV-TR, The Diagnostic and Statistical Manual of Mental Disorders, 4th edition, text revision; MMSE, Mini Mental State Examination; N/A, not available; NIA-AA, the National Institute of Aging –Alzheimer’s Association; NINCDS-ADRDA, the National Institute of Neurological and Communicative Disorders and Stroke-Alzheimer’s Disease and Related Disorders Association.

Risk of bias assessment and certainty of evidence

The risk of bias of the included studies is shown in Supplementary Figure 3. In a study by Swanson et al. the dropout rate was higher in the lecanemab group than in the placebo group (36% versus 23.7%), mostly due to ARIAs, which are drug-specific side effects. Therefore, we assessed “deviations from intended treatment”, “missing outcome data”, and “overall risk of bias” as some concerns in the Swanson et al. study. Similarly, in the Varela et al. study, the exercise group had a higher dropout rate than the control group, mostly due to non-adherence, so we assessed “deviations from intended treat” and “missing outcome data” and “overall risk of bias” as some concerns in the Varela et al. study. Publication bias was not assessed because the number of arms in each intervention did not exceed 10. Risk of bias, inconsistency, indirectness, and imprecision were assessed using the GRADE approach, all of which were rated as “not serious” and the level of certainty of evidence was rated as high.

Network meta-analysis results

Efficacy

The results of the primary analysis are shown in Table 2A. Short-term aerobic exercise and melatonin were significantly more effective than donanemab, lecanemab, aducanumab, and placebo. Inconsistency was not measured because no closed loop was formed in the network. The heterogeneity was high and statistically significant. In a prespecified sensitivity analysis excluding short-duration aerobic exercise, the significant heterogeneity disappeared and melatonin was significantly superior to lecanemab, aducanumab and placebo (Supplementary Table 2). As heterogeneity was not significant except for the exercise intervention, no further search for heterogeneity was conducted.

Table 2A

The differences in effect (SMD [95% confidence interval]) of donanemab, lecanemab, aducanumab, melatonin, and short-duration aerobic exercise on cognitive function in mild Alzheimer’s disease and mild cognitive impairment are represented in the net league table

Aerobic exercise for a short time
–0.40 [–1.76 : 0.96]	Melatonin
–2.13 [–3.32: –0.95]	–1.73 [–3.22: –0.25]	Donanemab
–2.25 [–3.36: –1.14]	–1.85 [–3.27: –0.42]	–0.12 [–1.38 : 1.15]	Lecanemab
–2.42 [–3.56: –1.27]	–2.02 [–3.47: –0.56]	–0.28 [–1.58 : 1.01]	–0.17 [–1.39 : 1.06]	Aducanumab
–2.68 [–3.40: –1.95]	–2.27 [–3.42: –1.12]	–0.54 [–1.48 : 0.40]	–0.43 [–1.27 : 0.42]	–0.26 [–1.15 : 0.63]	Placebo
I² = 70.1% [40.5% –85.0%]; Q = 26.74, p = 0.0008

Bolded values are statistically significant. Negative values mean that the column intervention is superior to the row intervention.

In a post-hoc sensitivity analysis with modified hierarchies, the results were generally similar to those in the primary analysis (Supplementary Table 3), and in the analysis with hierarchies in the order of MMSE and ADAS, exercise and lecanemab and melatonin and donanemab and lecanemab were not significantly different (Supplementary Table 4).

Tolerability

Donanemab, lecanemab and aducanumab were significantly less well tolerated than placebo (Table 2B). There were no significant differences in the other comparisons. The heterogeneity was not significant.

Table 2B

The relative tolerability (odds ratios) of donanemab, lecanemab, aducanumab, melatonin, and short-duration aerobic exercise are represented in the net league table

Aerobic exercise for a short time
0.95 [0.51 : 17.73]	Melatonin
0.52 [0.10 : 2.76]	0.55 [0.05 : 6.59]	Donanemab
0.83 [0.16 : 4.36]	0.88 [0.07 : 10.45]	1.60 [0.89 : 2.85]	Lecanemab
0.91 [0.17 : 4.87]	0.96 [0.08 : 11.58]	1.74 [0.91 : 3.33]	1.09 [0.58 : 2.03]	Aducanumab
1.69 [0.34 : 8.46]	1.78 [0.15 : 20.59]	3.24 [2.11 : 4.99]	2.03 [1.38 : 3.00]	1.87 [1.15 : 3.03]	Placebo
I² = 0% [0% –67.6%]; Q = 3.7, p = 0.81

Bolded values are statistically significant. Positive values mean that the column intervention is inferior to the row intervention.

Table 2C

The relative acceptability (odds ratios) of donanemab, lecanemab, aducanumab, melatonin, and short-duration aerobic exercise are represented in the net league table

Aerobic exercise for a short time
4.78 [0.66 : 34.37]	Melatonin
4.66 [1.00 : 21.66]	0.98 [0.25 : 3.81]	Donanemab
4.26 [0.92 : 19.71]	0.89 [0.23 : 3.46]	0.91 [0.52 : 1.59]	Lecanemab
4.95 [1.07 : 23.01]	1.04 [0.27 : 4.05]	1.06 [0.60 : 1.87]	1.16 [0.67 : 2.03]	Aducanumab
7.15 [1.62 : 31.51]	1.50 [0.41 : 5.50]	1.53 [1.03 : 2.28]	1.68 [1.14 : 2.47]	1.44 [0.97 : 2.15]	Placebo
I² = 37.5% [0% –72.4%]; Q = 11.2, p = 0.13

Bolded values are statistically significant. Positive values mean that the column intervention is inferior to the row intervention.

Acceptability

Short-term aerobic exercise was significantly less acceptable than donanemab, aducanumab and placebo. Donanemab and lecanemab were also significantly less acceptable than placebo without significant heterogeneity. There were no significant differences in the other comparisons.

DISCUSSION

This is the first network meta-analysis to evaluate the comparative efficacy of donanemab, lecanemab, aducanumab, melatonin, and short-term aerobic exercise on cognitive function in patients with mild AD and MCI. Short-term aerobic exercise and melatonin were significantly more effective than donanemab, lecanemab and aducanumab in the primary analysis, supporting our hypothesis. Although significant heterogeneity weakened the validity of the results of the primary efficacy analysis, melatonin was still significantly more effective than donanemab, lecanemab and aducanumab without significant heterogeneity in a sensitivity analysis excluding short-term aerobic exercise. In two further sensitivity analyses with different hierarchies of cognitive function measures, melatonin, and short-term aerobic exercise were as effective or more effective than donanemab, lecanemab, and aducanumab with significant heterogeneity. Although these results generally support the results of the primary efficacy analysis, the robustness of the better cognitive effects of short-term aerobic exercise is still low, as it was only shown in the analyses with significant heterogeneity. In addition, the intervention periods of the included studies regarding melatonin and aerobic exercise for a short time were a maximum of approximately 6 months, whereas those of aducanumab and lecanemab were as long as 18 months [26 , 45], indicating that short-duration aerobic exercise and melatonin take effect faster than lecanemab and aducanumab. Melatonin and short-term aerobic exercise were not significantly less tolerated than placebo, whereas donanemab, lecanemab, and aducanumab were significantly less tolerated than placebo, indirectly suggesting better tolerability of the formers. However, short-term aerobic exercise was significantly less acceptable than donanemab and aducanumab. The main reason for discontinuation of short-term aerobic exercise was non-adherence, suggesting that it may be difficult to maintain motivation even for a short time. Melatonin showed good acceptability with no significant difference from placebo, whereas donanemab and lecanemab showed significantly worse acceptability than placebo, indirectly suggesting that melatonin is more acceptable than donanemab, lecanemab. Considering a clinical usefulness in terms of efficacy, tolerability and acceptability, melatonin may be better than donanemab, lecanemab and aducanumab. Aerobic exercise for a short time might also be better than donanemab, lecanemab, and aducanumab if continued without drop-out due to non-adherence, as it is well tolerated and more effective, although less valid due to heterogeneity.

Although highly speculative, the reason for the superiority of exercise and melatonin over lecanemab and aducanumab may be that aducanumab and lecanemab only act on Aβ [26 , 41], whereas exercise and melatonin have broader targets, such as inhibition of abnormal phosphorylation of tau protein, anti-inflammatory effects, and production of BDNF [10 , 17–20]. Furthermore, since lecanemab and aducanumab were not significantly different from placebo in this meta-analysis and lithium, which prevents phosphorylation of Aβ and tau protein via inhibition of GSK-3β, was significantly better than aducanumab on cognitive function in mild AD and MCI in our previous meta-analysis [31], it may be necessary to target the aforementioned mechanisms to effectively treat mild AD and MCI. Furthermore, there is a discrepancy between the relatively inferior efficacy of antibody therapy shown in the present study and the solid brain amyloid clearance detected by amyloid PET in each of the included studies, suggesting that the latter may not correctly capture the removal of amyloid or the course of AD. This is supported by that amyloid PET is inferior to FDG-PET, in its short-term accuracy of progression and specificity in predicting conversion from MCI to AD [46] and supports the possibility, mentioned in the Introduction, that amyloid PET captures wandering where damage to the brain parenchyma results in the loss of neurons and supporting structures rather than the removal of Aβ by antibody therapy [3]. Prospective studies using FDG-PET at baseline and follow-up are therefore essential in the future, as this is the only valid way to demonstrate the efficacy of interventions to improve brain function.

The present study showed no statistical difference in cognitive function between melatonin and aerobic exercise for a short time, suggesting that melatonin is a good adaptation for people with sleep-wake rhythm disorders, and that aerobic exercise for a short time is a good adaptation for inactive people. It is important to note, however, that the effects of melatonin on cognitive function in mild AD and MCI may be approached both by treating sleep-wake rhythm disorders and by a different mechanism. This is because no studies of melatonin focused on sleep-wake disturbances as an inclusion criterion [43, 45]. Therefore, melatonin administration may be effective for cognitive function in people with mild AD and MCI without sleep-wake rhythm disorders, but further validation studies are needed.

Limitations

This study had several limitations. First, the simultaneous analysis of pharmacological and non-pharmacological treatments may violate the transitivity assumption, which is supported by the findings that heterogeneity was reduced and not significant in sensitivity analyses excluding exercise. Therefore, maximum caution should be taken in interpreting the results. One possible source of heterogeneity between pharmacotherapy and non-pharmacotherapy is that the exercise trials were not double-blind, which may have contributed to the improved effectiveness of exercise. Second, the number of studies and participants included were small, which can lead to cause both type I errors and type II errors. An accumulation of RCTs with direct comparisons between interventions is expected. Third, MCI is more likely to progress to AD, but not all MCI progresses to AD. However, as noted in the Introduction, the purpose of this study was to evaluate other interventions under the FDA indications for lecanemab and aducanumab. Forth, future studies examining long-term outcomes are needed because the results of this study were limited to a short period of time. Fifth, we were technically unable to perform the analysis using the three-level restricted maximum likelihood random effects model, but such a more accurate method of analysis is recommended for future studies.

Conclusions

In this study, we showed that melatonin may be more effective and tolerable than aducanumab and more effective, tolerable, and acceptable than donanemab, lecanemab on cognitive function in people with MCI and mild AD. Aerobic exercise for a short time might also be more effective than these anti-Aβ antibody therapies, although its validity is reduced by heterogeneity, with better tolerability and worse acceptability in this condition. In summary, melatonin may be better than donanemab, lecanemab, and aducanumab as a disease-modifying treatment for MCI and mild AD, while aerobic exercise for a short time also might be so if it can be done without dropping out due to non-adherence. To improve the validity of comparative effects, larger direct comparison RCTs need to be conducted to address issues of small sample size and heterogeneity. In such cases, assessment of brain function by FDG-PET is essential to ensure the validity of disease modifiability.

AUTHOR CONTRIBUTIONS

Itsuki Terao (Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Project administration; Resources; Software; Supervision; Validation; Visualization; Writing – original draft; Writing – review & editing); Wakako Kodama (Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Project administration; Resources; Software; Supervision; Validation; Visualization; Writing – original draft; Writing – review & editing).

Footnotes

ACKNOWLEDGMENTS

We would like to thank Editage () for English language editing.

FUNDING

The authors have no funding to report.

CONFLICT OF INTEREST

The authors have no conflict of interest to report.

DATA AVAILABILITY

The data that support the findings of this study are available from the corresponding author upon request.

The supplementary material is available in the electronic version of this article: .

References

Institute for Health Metrics and Evaluation , Grobal Health Data Exchange–https://vizhub.healthdata.org/gbd-results/.

Selkoe

, Hardy

(2016) The amyloid hypothesis of Alzheimer’s disease at 25 years. EMBO Mol Med 8, 595–608.

Hoilund-Carlsen

, Revheim

, Costa

, Alavi

, Kepp

, Sensi

, Perry

, Robakis

, Barrio

, Vissel

(2023) Passive Alzheimer’s immunotherapy: A promising or uncertain option? Ageing Res Rev 90, 101996.

Hampel

, Elhage

, Cho

, Apostolova

, Nicoll

JAR

, Atri

(2023) Amyloid-related imaging abnormalities (ARIA): Radiological, biological and clinical characteristics. Brain 146, 4414–4424.

Alves

, Kalinowski

, Ayton

(2023) Accelerated brain volume loss caused by anti-β-amyloid drugs: A systematic review and meta-analysis. Neurology 100, e2114–e2124.

Eisai,Eisai’s approach to U.S. pricing for Leqembi™(lecanemab), a treatment for early Alzheimer’s disease,sets forth our concept of“societal valueof medicine” in relation to “price of medicine”,https://www.eisai.com/news/2023/news202302.html.

Eisai, Biogen announces reduced price for Aduhelm® to improve access for patients with early Alzheimer’s disease in the United States,https://www.eisai.com/news/2021/news2021100.html.

Piller

(2022) Blots on a field? Science 377, 358–363.

Zhang

, Tian

, Wang

, Ma

, Tan

, Yu

(2021) The epidemiology of Alzheimer’s disease modifiable risk factors and prevention. J Prev Alzheimers Dis 8, 313–321.

10.

Gholipour

, Komaki

, Parsa

, Ramezani

(2022) Therapeutic effects of high-intensity interval training exercise alone and its combination with ecdysterone against amyloid beta-induced rat model of Alzheimer’s disease: A behavioral, biochemical, and histological study. Neurochem Res 47, 2090–2108.

11.

Yang

, Wu

, Li

, Dong

, Wu

, Lee

, Brann

, Lin

, Zhang

(2022) Long-term exercise pre-training attenuates Alzheimer’s disease-related pathology in a transgenic rat model of Alzheimer’s disease. Geroscience 44, 1457–1477.

12.

Liang

, Wang

, Zhang

, Huang

, Xie

, Chen

(2021) Exercise-induced benefits for Alzheimer’s disease by stimulating mitophagy and improving mitochondrial function. Front Aging Neurosci 13, 755665.

13.

Rody

, De Amorim

, De Felice

(2022) The emerging neuroprotective roles of exerkines in Alzheimer’s disease. Front Aging Neurosci 14, 965190.

14.

Zhang

, Zhen

, Su

, Chen

, Lv

, Yu

(2022) The effect of aerobic exercise on cognitive function in people with Alzheimer’s disease: A systematic review and meta-analysis of randomized controlled trials. Int J Environ Res Public Health 19, 15700.

15.

Lopez-Ortiz

, Valenzuela

, Seisdedos

, Morales

, Vega

, Castillo-Garcia

, Nistico

, Mercuri

, Lista

, Lucia

, Santos-Lozano

(2021) Exercise interventions in Alzheimer’s disease: A systematic review and meta-analysis of randomized controlled trials. Ageing Res Rev 72, 101479.

16.

Menczel Schrire

, Phillips

, Chapman

, Duffy

, Wong

, D’Rozario

, Comas

, Raisin

, Saini

, Gordon

, McKinnon

, Naismith

, Marshall

, Grunstein

, Hoyos

(2022) Safety of higher doses of melatonin in adults: A systematic review and meta-analysis. J Pineal Res 72, e12782.

17.

Matsubara

, Bryant-Thomas

, Pacheco Quinto

, Henry

, Poeggeler

, Herbert

, Cruz-Sanchez

, Chyan

, Smith

, Perry

, Shoji

, Abe

, Leone

, Grundke-Ikbal

, Wilson

, Ghiso

, Williams

, Refolo

, Pappolla

, Chain

, Neria

(2003) Melatonin increases survival and inhibits oxidative and amyloid pathology in a transgenic model of Alzheimer’s disease. J Neurochem 85, 1101–1108.

18.

Kamsrijai

, Wongchitrat

, Nopparat

, Satayavivad

, Govitrapong

(2020) Melatonin attenuates streptozotocin-induced Alzheimer-like features in hyperglycemic rats. Neurochem Int 132, 104601.

19.

Andrade

, Souza

, Azevedo

, Bail

, Zanata

, Andreatini

, Vital

(2023) Melatonin reduces β-amyloid accumulation and improves short-term memory in streptozotocin-induced sporadic Alzheimer’s disease model. IBRO Neurosci Rep 14, 264–272.

20.

Song

(2019) Pineal gland dysfunction in Alzheimer’s disease: Relationship with the immune-pineal axis, sleep disturbance, and neurogenesis. Mol Neurodegener 14, 28.

21.

Liu

, Zhou

, van Heerikhuize

, Hofman

, Swaab

(1999) Decreased melatonin levels in postmortem cerebrospinal fluid in relation to aging, Alzheimer’s disease, and apolipoprotein E-epsilon4/4 genotype. J Clin Endocrinol Metab 84, 323–327.

22.

Savaskan

, Ayoub

, Ravid

, Angeloni

, Fraschini

, Meier

, Eckert

, Müller-Spahn

, Jockers

(2005) Reduced hippocampal MT2 melatonin receptor expression in Alzheimer’s disease. J Pineal Res 38, 10–16.

23.

Brunner

, Sözer-Topcular

, Jockers

, Ravid

, Angeloni

, Fraschini

, Eckert

, Müller-Spahn

, Savaskan

(2006) Pineal and cortical melatonin receptors MT1 and MT2 are decreased in Alzheimer’s disease. Eur J Histochem 50, 311–316.

24.

Sumsuzzman

, Choi

, Jin

, Hong

(2021) Neurocognitive effects of melatonin treatment in healthy adults and individuals with Alzheimer’s disease and insomnia: A systematic review and meta-analysis of randomized controlled trials. Neurosci Biobehav Rev 127, 459–473.

25.

Hutton

, Salanti

, Caldwell

, Chaimani

, Schmid

, Cameron

, Ioannidis

, Straus

, Thorlund

, Jansen

, Mulrow

, Catalá-López

, Gøtzsche

, Dickersin

, Boutron

, Altman

, Moher

(2015) The PRISMA extension statement for reporting of systematic reviews incorporating network meta-analyses of health care interventions: Checklist and explanations. Ann Intern Med 162, 777–784.

26.

Budd Haeberlein

, Aisen

, Barkhof

, Chalkias

, Chen

, Cohen

, Dent

, Hansson

, Harrison

, von Hehn

, Iwatsubo

, Mallinckrodt

, Mummery

, Muralidharan

, Nestorov

, Nisenbaum

, Rajagovindan

, Skordos

, Tian

, van Dyck

, Vellas

, Wu

, Zhu

, Sandrock

(2022) Two randomized phase 3 studies of aducanumab in early Alzheimer’s disease. J Prev Alzheimers Dis 9, 197–210.

27.

van Dyck

, Swanson

, Aisen

, Bateman

, Chen

, Gee

, Kanekiyo

, Li

, Reyderman

, Cohen

, Froelich

, Katayama

, Sabbagh

, Vellas

, Watson

, Dhadda

, Irizarry

, Kramer

, Iwatsubo

(2023) Lecanemab in early Alzheimer’s disease. N Engl J Med 388, 9–21.

28.

Sterne

JAC

, Savović

, Page

, Elbers

, Blencowe

, Boutron

, Cates

, Cheng

, Corbett

, Eldridge

, Emberson

, Hernán

, Hopewell

, Hróbjartsson

, Junqueira

, Jüni

, Kirkham

, Lasserson

, Li

, McAleenan

, Reeves

, Shepperd

, Shrier

, Stewart

, Tilling

, White

, Whiting

, Higgins

JPT

(2019) RoB 2: A revised tool for assessing risk of bias in randomised trials. BMJ 366, 4898.

29.

Guyatt

, Oxman

, Vist

, Kunz

, Falck-Ytter

, Alonso-Coello

, Schünemann

(2008) GRADE: An emerging consensus on rating quality of evidence and strength of recommendations. BMJ 336, 924–926.

30.

Higgins

, Thomas

, Chandler

, Cumpston

, LI

, Pag

, Welch

, Cochrane Handbook for Systematic Reviews of Interventions version 6.3 (updated February 2022),Cochrane,http://www.training.cochrane.org/handbook.

31.

Terao

, Honyashiki

, Inoue

(2022) Comparative efficacy of lithium and aducanumab for cognitive decline in patients with mild cognitive impairment or Alzheimer’s disease: A systematic review and network meta-analysis. Ageing Res Rev 81, 101709.

32.

Rücker

G KU

, König

, Efthimiou

, Davies

, Papakonstantinou

, Schwarzer

,netmeta: Network Meta-Analysis using Frequentist Methods . R package version 2.5-0,https://CRAN.R-project.org/package=netmeta

33.

R Core TeamR: A Language and Environment for Statistical Computing, R Foundation for Statistical Computing,https://www.R-project.org/

34.

Higgins

, Thomas

, Chandler

, Cumpston

, LI

, Pag

, Welch

.Cochrane Handbook for Systematic Reviews of Interventions version 6.4 (updated August 2023),http://www.training.cochrane.org/handbook

35.

Jackson

, Barrett

, Rice

, White

, Higgins

(2014) A design-by-treatment interaction model for network meta-analysis with random inconsistency effects. Stat Med 33, 3639–3654.

36.

Efthimiou

, Debray

, van Valkenhoef

, Trelle

, Panayidou

, Moons

, Reitsma

, Shang

, Salanti

(2016) GetReal in network meta-analysis: A review of the methodology. Res Synth Methods 7, 236–263.

37.

Arcoverde

, Deslandes

, Moraes

, Almeida

, Araujo

, Vasques

, Silveira

, Laks

(2014) Treadmill training as an augmentation treatment for Alzheimer’s disease: A pilot randomized controlled study. Arq Neuropsiquiatr 72, 190–196.

38.

Enette

, Vogel

, Merle

, Valard-Guiguet

, Ozier-Lafontaine

, Neviere

, Leuly-Joncart

, Fanon

, Lang

(2020) Effect of 9 weeks continuous vs. interval aerobic training on plasma BDNF levels, aerobic fitness, cognitive capacity and quality of life among seniors with mild to moderate Alzheimer’s disease: A randomized controlled trial. Eur Rev Aging Phys Act 17, 2.

39.

Mintun

, Lo

, Duggan Evans

, Wessels

, Ardayfio

, Andersen

, Shcherbinin

, Sparks

, Sims

, Brys

, Apostolova

, Salloway

, Skovronsky

(2021) Donanemab in early Alzheimer’s disease. N Engl J Med 384, 1691–1704.

40.

Sims

, Zimmer

, Evans

, Lu

, Ardayfio

, Sparks

, Wessels

, Shcherbinin

, Wang

, Monkul Nery

, Collins

, Solomon

, Salloway

, Apostolova

, Hansson

, Ritchie

, Brooks

, Mintun

, Skovronsky

, TRAILBLAZER-ALZ 2 Investigators(2023) Donanemab in early symptomatic Alzheimer disease: The TRAILBLAZER-ALZ 2 randomized clinical trial. JAMA 330, 512–527.

41.

Swanson

, Zhang

, Dhadda

, Wang

, Kaplow

, Lai

RYK

, Lannfelt

, Bradley

, Rabe

, Koyama

, Reyderman

, Berry

, Gordon

, Kramer

, Cummings

(2021) A randomized, double-blind, phase 2b proof-of-concept clinical trial in early Alzheimer’s disease with lecanemab, an anti-Abeta protofibril antibody. Alzheimers Res Ther 13, 80.

42.

Varela

, Ayan

, Cancela

, Martin

(2012) Effects of two different intensities of aerobic exercise on elderly people with mild cognitive impairment: A randomized pilot study. Clin Rehabil 26, 442–450.

43.

Wade

, Farmer

, Harari

, Fund

, Laudon

, Nir

, Frydman-Marom

, Zisapel

(2014) Add-on prolonged-release melatonin for cognitive function and sleep in mild to moderate Alzheimer’s disease: A 6-month, randomized, placebo-controlled, multicenter trial. Clin Interv Aging 9, 947–961.

44.

, Yu

, Sun

, Hu

, Geng

(2020) Dietary melatonin therapy alleviates the lamina cribrosa damages in patients with mild cognitive impairments: A double-blinded, randomized controlled study. Med Sci Monit 26, e923232.

45.

, Wang

L-L

, Dammer

, Li

C-B

, Xu

, Chen

S-D

, Wang

(2015) Melatonin for sleep disorders and cognition in dementia. Am J Alzheimers Dis Other Demen 30, 439–447.

46.

Chételat

, Arbizu

, Barthel

, Garibotto

, Law

, Morbelli

, van de Giessen

, Agosta

, Barkhof

, Brooks

, Carrillo

, Dubois

, Fjell

, Frisoni

, Hansson

, Herholz

, Hutton

, Jack

Jr. , Lammertsma

, Landau

, Minoshima

, Nobili

, Nordberg

, Ossenkoppele

, Oyen

WJG

, Perani

, Rabinovici

, Scheltens

, Villemagne

, Zetterberg

, Drzezga

(2020) Amyloid-PET and (18)F-FDG-PET in the diagnostic investigation of Alzheimer’s disease and other dementias. Lancet Neurol 19, 951–962.