Drug development in Alzheimer's disease supported by translational cerebrospinal fluid biomarkers
7 December 2018
UAntwerp, Campus Drie Eiken, Building R, Auditorium R1 - Universiteitsplein 1 - 2610 Wilrijk (Antwerp) (route: UAntwerpen, Campus Drie Eiken
4:00 PM - 6:00 PM
Sebastiaan Engelborghs, Johannes Streffer
PhD defence Maarten Timmers - Department of Biomedical Sciences
In this thesis an optimized protocol for frequent cerebrospinal fluid (CSF) sampling using a spinal catheter was defined and applied during the clinical development of the β-secretase inhibitor, atabecestat, targeting amyloid-β (Aβ) as a potential treatment for Alzheimer’s disease (AD), to evaluate and predict the treatment effect of atabecestat and to study translational biomarkers in the field of AD.
Frequent CSF sampling using a spinal catheter is hampered by pronounced increases in Aβ limiting its use in detecting treatment effects or evaluating pharmacokinetic (PK)/pharmacodynamic (PD) characteristics of Aβ-directed treatments. Our findings confirmed that CSF Aβ levels are substantially affected by sampling frequency. Adjusting the sampling scheme from a high frequency to a low frequency with lower collection volumes reduced this influence on Aβ levels to rather non-significant levels.
Frequent CSF sampling (by spinal catheter or frequent lumbar punctures) was successfully applied during the clinical development of atabecestat. Atabecestat showed strong dose dependent peripheral and central Aβ lowering effects (up to 95%) in healthy elderly participants. The observed effects on Aβ combined with reported dose-related changes in CSF levels for the amyloid precursor protein (APP) fragments, i.e., sAPPα and sAPPβ, their magnitude and direction of change, are consistent with the mode of action of atabecestat. Atabecestat inhibits APP cleavage by b-secretase and subsequent Aβ production.
Frequent CSF sampling enabled detailed PK/PD profiling of atabecestat in healthy participants and the development of a PK/PD model to predict treatment effects in AD. PK/PD model simulations based on healthy participant data were confirmed in a 4-week study in early AD showing that qd dosing with 10 mg and 50 mg atabecestat attains 60% to 70% and 90% Aβ reduction. The effects on APP fragments (sAPPα and sAPPβ) observed in healthy participants were confirmed in early AD. These confirmatory data allow prediction of Aβ reduction for atabecestat doses, independent of disease stage or population.
Genetic evidence points towards Aβ as the cause of AD, however the AD clinical syndrome appears more closely related to the progression of tau pathology. An Aβ-dependent correlation was shown between cognitive performance and CSF tau suggestive of an interaction between both neuropathological hallmarks with Aβ being the initiating step and tau driving cognitive impairment. A finding that may shift the amyloid cascade hypothesis towards an interaction model.
Overall the findings described in this thesis emphasize the importance of CSF markers in understanding AD, its progression and the impact of treatment.