# Target Engagement and Cognitive Outcomes of Anti‐Amyloid Immunotherapies in Early‐Onset Alzheimer's Disease: First Report from the LEADS Study

## TL;DR

This study examines how anti-amyloid treatments affect brain pathology and cognitive decline in early-onset Alzheimer's patients, finding strong amyloid clearance but no clear cognitive benefits.

## Contribution

The study provides real-world evidence of anti-amyloid immunotherapies' effects in early-onset Alzheimer's using longitudinal biomarker and cognitive data.

## Key findings

- Treated patients showed significant decreases in Aβ burden post-treatment.
- No significant changes in tau burden or cognitive scores were observed.
- Observational data complements randomized trials in evaluating treatment efficacy.

## Abstract

In clinical trials, monoclonal antibodies targeting Aβ pathology have shown strong target engagement, resulting in rapid Aβ clearance and a deceleration in rate of clinical decline. Now that these treatments are approved and implemented in clinical practice, we could assess their effects in observational studies involving these patients.

We analyzed data from 20 participants with early‐onset Alzheimer's disease (EOAD) in the Longitudinal Early‐Onset Alzheimer's Disease Study (LEADS) cohort, treated with Aducanumab (n = 4), Lecanemab (n = 15), or both (one transitioning from Aducanumab to Lecanemab). All participants had MCI or mild dementia at baseline, longitudinal Aβ and tau PET, as well as cognitive assessments, with at least one observation before and after treatment initiation. We applied piecewise regression with a knot at the treatment start, to evaluate changes in Aβ and tau PET burden and Clinical Dementia Rating‐Sum of Boxes (CDR‐SB) scores. We compared the trajectories of treated participants with an untreated group (i.e., treated‐untreated comparison) from LEADS, matched for age, sex, APOE ε4 genotype, pretreatment Aβ and tau PET load, CDR‐SB, and follow‐up duration, using a 1:3 matching design.

The median treatment duration was 8 months (IQR=5‐10). In the piecewise regression model, the treated group showed significant decreases in Aβ burden post‐treatment (Δ=‐52 Centiloids/yr, p <0.001) with widespread neocortical involvement (Figure 1). However, no significant inflection in tau burden (Δ=0 SUVR/yr, p = 0.58) or CDR‐SB (Δ=0.3 units/yr, p = 0.57) trajectories was observed. In the treated‐untreated comparison, the treated group showed a trend toward slower increases in CDR‐SB scores post‐treatment (ΔT=‐1.8, p = 0.07) compared to the untreated group (Figure 2). Aβ levels significantly decreased in the treated group compared to the untreated group (ΔΤ=‐8.5, p <0.001). No significant differences in tau trajectories were observed between groups (ΔT=0.4, p = 0.68), with both showing increases in cortical regions of interest.

We observed excellent target engagement, with piecewise regression models showing rates of Aβ clearance comparable to those seen in Phase 3 trials. The study was underpowered to detect cognitive benefits, which are limited over a short follow‐up interval. These findings underscore the utility of observational studies with biomarker data in evaluating treatment efficacy, offering insights that complement traditional randomized trials.