Re‐estimation of drug‐specific amyloid removal parameters and the rate constant for pathogenic tau turnover brings the Q‐ATN model into better alignment with recent phase 3 data from gantenerumab and lecanemab.

Background: The Q‐ATN model (Mazer et al., Alzheimer's & Dementia. 2022; https://doi.org/10.1002/alz/12877) is a quantitative semi‐mechanistic representation of the A (amyloid)/T (tau)/N (neurodegeneration) framework for simulating the natural history of Alzheimer's disease and the effects of anti‐amyloid treatment. The recently reported phase 3 trials of gantenerumab (GRADUATE 1&2) and lecanemab (CLARITY) provide additional data to update model parameters and assess the assumptions of the model. Method: The drug‐specific amyloid removal parameters (αrem) for gantenerumab and lecanemab were re‐estimated from the mean amyloid PET data in GRADUATE 1&2 and CLARITY. Using the new αrem values, Q‐ATN simulations of the Clinical Dementia Rating – Sum of Boxes (CDR‐SB) and tau PET trajectories were made for each study. The rate constant for pathogenic tau turnover (ktau), which was uncertain due to limited data, was re‐estimated using a sensitivity analysis of the CDR‐SB treatment effects. Results from the aducanumab EMERGE and ENGAGE studies were also included in the analysis. Result: The simulated increase in amyloid PET agreed well with the mean data from the placebo groups. For GRADUATE 1&2, αrem was 34% and 41% lower, respectively, than the prior value (0.0137 Yr−1/(ug/mL)) estimated from the gantenerumab open‐label extension studies. For CLARITY, αrem was 26% lower than the prior value (0.0109 Yr−1/(ug/mL)) estimated from the lecanemab phase 2 study. For GRADUATE 1&2, the simulated CDR‐SB trajectories for the placebo groups were in good agreement with the observed increases from baseline. For CLARITY, the simulation overestimated the observed increase. The re‐estimated value of ktau (0.2 Yr−1), which corresponds to a 3.5 Yr half‐life, yielded treatment effect sizes on CDR‐SB within the 95% confidence intervals for all treatment arms (N = 7). The prior value (0.5 Yr−1) yielded treatment effect sizes that were too large in three cases (see Figure). Tau PET simulations (with ktau = 0.2 Yr−1) were consistent with medial temporal ROI data from lecanemab and aducanumab; tau PET data from gantenerumab are pending. Conclusion: The updated αrem and ktau parameters bring the Q‐ATN model into better alignment with recent phase 3 data. Forthcoming data from the Donanemab TRAILBLAZER‐ALZ 2 study could further help assess the model assumptions. [ABSTRACT FROM AUTHOR]