Efficacy and safety of passive immunotherapies targeting amyloid beta in Alzheimer’s disease: A systematic review and meta-analysis

Background While recently U.S. FDA-approved anti-amyloid beta (anti-Aβ) monoclonal antibodies (mAbs) offer new treatment approaches for patients suffering from Alzheimer’s disease (AD), these medications also carry potential safety concerns and uncertainty about their efficacy for improving cognitive function. This study presents an updated meta-analysis of cognitive outcomes and side effects of anti-Aβ mAbs from phase III randomized controlled trials (RCTs) in patients with sporadic AD. Methods and Findings Phase III randomized, placebo-controlled blinded trial evaluating the efficacy and safety of anti-Aβ mAbs in patients with AD were identified through a search in clinical trials.gov, PubMed and Embase on January 14th, 2024. The retrieved studies were further screened from January 15th, 2024, to February 14th, 2024. We included studies that had been published in any language. Quality of trials was assessed using the Jadad score and publication bias was assessed using Egger’s test and Funnel plot. Primary outcomes were mean changes from baseline to post-treatment in Clinical Dementia Rating scale-Sum of Boxes (CDR-SB) and AD Assessment Scale-Cognitive Subscale (ADAS-Cog) scores, and secondary outcomes were adverse events including amyloid-related imaging abnormalities with edema (ARIA-E), and ARIA with hemorrhage (ARIA-H). Random-effects meta-analysis and meta-regression analyses were conducted. The literature search identified 13 phase III RCTs, which included 18,826 patients with mild cognitive impairment or dementia due to AD. Compared with placebo, treatment with mAbs significantly improved cognitive performance on CDR-SB (mean difference −0.25, 95% confidence interval [CI] [−0.38, −0.11]) and ADAS-Cog (standardized mean difference −0.09, 95% CI [−0.12, −0.06]), in which a negative change indicates improvement for both scores. Meta-regression analysis suggested that trials enrolling patients with early-stage AD were associated with better efficacy. Elevated risk of ARIA-E (risk ratio [RR] 9.79, 95% CI [5.32,18.01]), ARIA-H (RR 1.94, 95% CI [1.47,2.57]), and headaches (RR 1.21, 95% CI [1.10,1.32]) were noted. Statistical heterogeneity was relatively high for ARIA-E and ARIA-H, leading to wide confidence intervals and considerable variability in effect sizes, though meta-regression was conducted to address this. Furthermore, differences in trial designs introduce limitations in cross-trial comparisons. Conclusions Anti-Aβ mAb therapy slows cognitive decline, but with small effect sizes, and raises potential concerns about ARIA and headaches.