Alzheimer disease is a progressive, irreversible neurodegenerative disorder that affects up to 50 percent of adults by age 85. Accumulation of amyloid beta peptide in the brain is thought to play a key role and has been a target of clinical development for more than 25 years. The apolipoprotein E 4 (APOE4) gene is thought to increase the accumulation of amyloid and be a risk factor for late onset Alzheimer disease. The difficult and fatal course of Alzheimer disease, paired with limited efficacy of available treatments, breeds high hopes for a miracle drug. Aducanumab, a human monoclonal antibody directed against amyloid beta, was granted accelerated FDA approval in June 2021. On the heels of a phase 1b study (PRIME) that suggested slowing of clinical decline, investigators at Biogen designed two identical trials to investigate the efficacy and safety of aducanumab.

Investigators enrolled 1,647 (ENGAGE) and 1,638 (EMERGE) participants aged 50-85 years old with Alzheimer disease implied by amyloid deposits seen on PET scans in two identically designed randomized phase 3 clinical trials. Participants were randomized 1:1:1 to receive high-dose aducanumab (6-10 mg/kg), low-dose aducanumab (3-6 mg/kg), or placebo, all administered IV every four weeks for 76 weeks. Because of concerns for amyloid-related imaging abnormalities (ARIA) that can occur at high doses of aducanumab and in APOE4 carriers, participants identified as APOE4 carriers received dosages at the lower end of the ranges for both high- and low-dose aducanumab (6 and 3 mg/kg, respectively). ARIA can manifest clinically as headaches, vomiting, confusion, and gait disturbance. The primary endpoint was change from baseline to week 78 in the Clinical Dementia Rating Sum of Boxes (CDR-SB) score. This zero-18-point scale assesses both cognition and function, with higher scores indicating more severe impairment. Secondary endpoints included change from baseline on several other clinical scales, adverse events, and CSF biomarkers. Brain MRIs were planned for all participants to assess for ARIA at screening and at seven additional time points (26,280 total MRIs). There were a number of protocol amendments due to failure of lower doses of other agents and recognition of a relatively benign course in most ARIA cases. Amendments included an increase in the target aducanumab dose from six to 10 mg/kg for APOE4 carriers in the high-dose aducanumab group and more leniency around continued dose titration in participants who developed ARIA. Both trials were stopped early due to prespecified futility criteria being met at a planned interim analysis, at which point 53 percent of EMERGE participants and 57 percent of ENGAGE participants had completed the study.

Results from the two identically designed trials were discordant. For patients who had received high-dose aducanumab compared to placebo, data from the EMERGE trial demonstrated a small but significant reduction in CDR-SB score (-0.39 points [95% CI -0.69 to -0.09]), but the ENGAGE trial demonstrated a nonsignificant increase in CDR-SB score (0.03 points, 95% CI -0.26 to +0.33). Neither trial found significant differences in the primary or secondary clinical outcomes comparing low-dose aducanumab to placebo. However, both trials found a dose-dependent increase in adverse events, with ARIA-edema in 43 percent (42 percent) of APOE4 carriers receiving high-dose aducanumab, 30 percent (29 percent) of those receiving low-dose, and two percent (2 percent) of those receiving placebo in the EMERGE and (ENGAGE) trials, respectively.

The bottom line here is that a promising new treatment for Alzheimer disease had a disappointing show in two expensive clinical trials. The effort spent on extensive post hoc exploratory analyses trying to explain the lack of statistical significance and discordance between trials does not offer us much confidence that there is efficacy to be found for aducanumab. What’s clear, however, is that higher doses result in higher risk of harm. The investigators (and Biogen) wanted this drug to work – we all do. But unless we have reason to believe that at some higher dose the dose-response curve for benefit will suddenly accelerate and outpace the dose-response curve for harm, further trials could be considered unethical. Unfortunately, it seems that we have a futile attempt at a new treatment for a disease that could desperately use a miracle.

Read the initial reaction from DynaMed Deputy Editor of Neurology in this EBSCO Health Notes blog article.