Modeling amyloid plaque turnover dynamics improves characterization of drug effects

Effect on amyloid plaque as measured by positron emission tomography imaging with Centiloid standardization of two therapeutic approaches targeting amyloid beta (Aβ) was investigated using exposure-response modeling.

Individual-level verubecestat data from the APECS trial were pooled with summary-level data from the literature for amyloid monoclonal antibodies (mAbs) and fitted in a joint non-linear mixed-effects model.

An indirect-response (turnover) model with verubecestat inhibiting plaque formation and mAbs stimulating plaque removal well represented the data. The estimated plaque elimination half-life was 6.4 years. Daily verubecestat 40 mg was estimated to reduce formation by 91.8%. Aducanumab 10 mg/kg every 4 weeks (Q4W), donanemab 1400 mg Q4W, gantenerumab 1200 mg Q4W, and lecanemab 10 mg/kg Q2W were estimated to increase the removal rate by 9.3-, 18.6-, 5.3-, and 13.8-fold, respectively.

The model provides a fundamental measure of drug effects on plaque, independent of disease stage and study-design factors, improving cross-study comparisons and enabling predictions.

The plaque turnover model describes natural progression and BACE and mAb intervention.The model estimation of the underlying plaque elimination half-life is 6.4 years.Approach improves cross-study comparison independently of population and study design.Predictions of alternative regimens/therapeutic approaches will aid future study design.