Evaluating amyloid-beta as a surrogate endpoint in trials of anti-amyloid-beta drugs in Alzheimer’s disease: a Bayesian meta-analysis

Aim: The use of amyloid-beta (Aβ) clearance to support regulatory approvals of drugs in Alzheimer’s disease (AD) remains controversial. We evaluate Aβ as a potential trial-level surrogate endpoint for clinical function in AD.

Materials & methods: Data on the effectiveness of anti-Aβ monoclonal antibodies (MABs) on Aβ and multiple clinical outcomes were identified from randomized controlled trials through a literature review. A Bayesian bivariate meta-analysis was used to evaluate Aβ as a surrogate endpoint for clinical function across all MABs and for each individual anti-Aβ MAB. The analysis for individual therapies was conducted in subgroups of treatments and by applying Bayesian hierarchical models to borrow information across treatments.

Results: We identified 23 randomized controlled trials with 39 treatment contrasts for seven MABs. The surrogate relationship between treatment effects on Aβ and Clinical Dementia Rating-Sum of Boxes (CDR-SOB) across all MABs was strong: with a meaningful slope of 1.41 (0.60, 2.21) and small variance of 0.02 (0.00, 0.05). For individual treatments, the surrogate relationships were suboptimal, displaying large uncertainty. Sharing information across treatments considerably reduced the uncertainty, resulting in moderate surrogate relationships for aducanumab and lecanemab. No meaningful association was detected for other clinical outcomes, including Mini Mental State Examination and Alzheimer’s Disease Assessment Scale-Cognitive Subscale.

Conclusion: Although our results from the analysis of data across all MABs suggested that Aβ was a potential surrogate endpoint for CDR-SOB, individually the surrogacy patterns varied across treatments and showed no evidence of association. Bayesian information-sharing revealed moderate surrogate relationship only for aducanumab and lecanemab.