Modeling and Simulation to Inform Potential New Strategies for BACEi in Treatment of Alzheimers Disease

Abstract

AbstractBackgroundThe β‐secretase‐1 inhibitors (BACEi), including verubecestat, were extensively studied in prodromal to moderate AD and demonstrated early cognitive decline at doses achieving >50% inhibition of amyloid production. Questions remain as to whether BACEi may still have utility, if used earlier in disease, in combination with other therapies, and/or at lower doses. Modeling and simulation approaches were used to maximize learnings from the prior trials and to make predictions around potential alternative therapeutic approaches.MethodTwo models were developed: 1) A joint plaque turnover model, linking drug exposure with amyloid plaque load (PET data in centiloids), to describe natural progression and treatment response to BACEi and amyloid mAb therapies. All plaque data from natural progression (control arms), BACEi and amyloid mAb treatment arms were well described by a joint model, after accounting for differences in baseline plaque load, dosing and titration regimens, and mechanism of action. 2) Disease progression models for the clinical efficacy outcomes (ADAS‐cog11, CDR‐SB) were developed using a methodology similar to Ref. Evidence for BACEi drug effects (on progression rate and/or early detrimental offset) was tested and dependencies on dose, drug exposure and patient characteristics were explored.ResultThe joint plaque turnover model well described natural progression and treatment response to BACEi and amyloid mAb and allows for prediction of alternative regimens and therapeutic approaches. Simulations indicate that 20‐30% inhibition of BACE would be sufficient to stabilize plaque loads in prodromal patients. Disease progression results for ADAS‐cog11 and CDR‐SB identified a verubecestat drug effect of early detrimental offset with no effect on rate of disease progression supported. The magnitude of the early detrimental offset was small relative to the overall progression seen over 18‐24 months and results suggested that this effect may be reduced in advanced AD patients. A simulation engine utilizing the progression model allows for estimation of sample size needed to check for the detrimental effect in future trials.ConclusionModel‐based simulations can help inform future BACEi therapeutic strategies and trial designs. Reference: Clin Pharmacol Ther. 2015 Mar;97(3):210‐4