Donanemab exposure-response in early symptomatic Alzheimer's disease

Abstract

Introduction: These analyses aimed to identify factors impacting donanemab exposure, amyloid plaque, and clinical efficacy in early symptomatic Alzheimer's disease using a population pharmacokinetic/pharmacodynamic (PK/PD) approach.

Methods: Analyses included donanemab trial participants (NCT02624778; NCT03367403; NCT04640077; NCT04437511). Dose- and exposure-response relationships were characterized relative to amyloid plaque lowering using indirect response PK/PD and disease progression models.

Results: Donanemab population PK was described by a biphasic distribution with an estimated terminal elimination half-life of approximately 12.1 days for a typical participant (72 kg body weight, 1:2560 maximum antidrug antibody [ADA] titer). Amyloid reduction was associated with maintaining median serum donanemab concentrations over 15 µg/mL (95% confidence interval: 8.54, 18.0). After completing active treatment, simulations showed an estimated plaque reaccumulation rate (median, 95% confidence interval) of 2.8 (2.16, 3.11) Centiloids/year. The donanemab disease progression model showed a clear treatment effect.

Discussion: These donanemab models can inform dosing strategies in future clinical trials.

Highlights: Weight-based and flat dosing had similar exposure metrics; flat dosing was adopted. Donanemab exposure was influenced by weight and titer (not clinically relevant). 2.8 Centiloids/year amyloid reaccumulation rate observed upon donanemab treatment completion. Around 30% reduction in disease progression rate on treatment for integrated Alzheimer's Disease Rating Scale (iADRS) and Clinical Dementia Rating Scale-Sum of Boxes (CDR-SB).

Keywords: Alzheimer's disease; CDR‐SB; amyloid plaque; cognitive/functional endpoint; disease progression; donanemab; exposure; iADRS; nonlinear mixed effects modeling.

FIGURE 1

Effect of covariates on donanemab exposure. (A) Impact of body weight on AUC and C_max at steady state. Expressed as donanemab exposures relative to a reference participant (50th percentile body weight [72 kg] with titers sampled from observed). Body weight categories of 48 and 104 kg represent the 5% and 95% percentiles of the PK analysis set, respectively. To illustrate the body weight effect, 1500 virtual participants were simulated (n = 500 in each weight category); median (95% CIs) relative to the reference participant for each weight category for AUC and C_max are shown. (B) Impact of ADA titer on AUC and C_max at steady state. Time‐varying ADA titer (at maximum effect, Week 24) was summarized from the observed time course of titers over time. Donanemab clearance and AUC were calculated at the median, 5th/95th percentiles of the titer time course. Plot displays relative change in a parameter with 5th to 95th percentiles relative to the reference (most common, denoted as low on the graph), with low titer time course and the sample's body weight. Filled circle and triangle: median ± 90% CI of AUC and C_max, vertical dashed line: reference; shaded gray area represented 30% difference in exposure: 0.7–1.43. ADA, antidrug antibody; AUC, area under the curve; CI, confidence interval; C_max, maximum concentration; PK, pharmacokinetics.

FIGURE 2

Body weight impact using simulations with the final population PK model. Fixed dosing versus body weight‐based dosing was compared using a simulation study with the final population PK model for maximum concentration (A), trough concentration (B), average drug concentration (C), and exposure (AUC; D) at steady state. Similar to the approach by Wang et al., the variability of exposure after both dosing approaches was examined. Lower and upper hinges correspond to 25th and 75th percentiles; the upper whisker extends from the hinge to the largest value no further than 1.5 * interquartile range. AUC, area under the concentration versus time curve; C_avg, average drug concentration; C_max, maximum observed drug concentration; C_trough, drug concentration before the next dose; PK, pharmacokinetics; Q4W, every 4 weeks.

FIGURE 3

Observed individual participant amyloid plaque trajectories used with the exposure‐amyloid plaque model. Observed individual amyloid plaque trajectories (dashed lines) from 1022 participants receiving donanemab treatment and 1001 receiving placebo. Q4W, every 4 weeks.

FIGURE 4

Amyloid plaque level over time using treatment exposure‐amyloid plaque reduction model. Amyloid plaque level over time using treatment exposure‐response model, stratified by participants achieving less than 11 Centiloids at Week 24. (A) Duration of 1.5 years, supported by data from TRAILBLAZER‐ALZ and TRAILBLAZER‐ALZ 2. (B) Duration of 5 years, supported by data from TRAILBLAZER‐EXT, Part C (plotted with observed data [mean ± SD]). (C) Duration extrapolated beyond the observed data range to 15 years. In A and B, participants (N = 10,000) were simulated to follow the dosing treatment regimen as described in Section 2. The model was developed using data from the phase 1 study, TRAILBLAZER‐ALZ, TRAILBLAZER‐ALZ 2, and TRAILBLAZER‐ALZ‐EXT. Dashed reference line shows 24.1 Centiloids. Median (90% prediction intervals) are represented by solid line and shaded area. N, number of participants.

FIGURE 5

Time to achieve amyloid plaque clearance stratified by baseline amyloid level. Percentage of participants achieving amyloid plaque clearance (< 24.1 Centiloids) by duration on treatment and by quartiles of baseline amyloid PET, assessed using the treatment exposure‐response model. Median (95% confidence interval) is represented by the solid line and shaded area. PET, positron emission tomography; Q, quartiles.

FIGURE 6

Disease progression model‐predicted change from baseline for the overall tau population. Disease progression model‐predicted change from baseline on iADRS (A) and CDR‐SB (B). Participants were simulated to follow the dosing treatment regimen, including the potential for down‐titration based on amyloid plaque level at Weeks 24 and 52, as described in Section 2. Median (95% confidence interval) is shown by the solid line and shaded areas. CDR‐SB, Clinical Dementia Rating Scale‐Sum of Boxes; iADRS, integrated Alzheimer's Disease Rating Scale.

FIGURE 7

Predicted percentage of participants in the TRAILBLAZER‐ALZ 2 safety addendum with amyloid plaque clearance over time and by cumulative donanemab dose. Using information on baseline amyloid levels and the established exposure‐amyloid plaque model^a, 1000 virtual participants across 100 studies were simulated with a percentage of participants achieving < 24.1 Centiloids. Median (95% CI) is shown by the solid line and shaded areas. ^aExposure‐amyloid plaque model did not include TRAILBLAZER‐ALZ 2 long‐term extension data. CI, confidence interval; PET, positron emission tomography.