Alzheimers Dement. 2023 May 8. doi: 10.1002/alz.13108. Online ahead of print.

ABSTRACT

We propose an unbiased methodology to rank compounds for advancement into comprehensive preclinical testing for Alzheimer’s disease (AD). Translation of compounds to the clinic in AD has been hampered by poor predictive validity of models, compounds with limited pharmaceutical properties, and studies that lack rigor. To overcome this, MODEL-AD’s Preclinical Testing Core developed a standardized pipeline for assessing efficacy in AD mouse models. We hypothesize that rank-ordering compounds based upon pharmacokinetic, efficacy, and toxicity properties in preclinical models will enhance successful translation to the clinic. Previously compound selection was based solely on physiochemical properties, with arbitrary cutoff limits, making ranking challenging. Since no gold standard exists for systematic prioritization, validating a selection criteria has remained elusive. The STOP-AD framework evaluates the drug-like properties to rank compounds for in vivo studies, and uses an unbiased approach that overcomes the validation limitation by performing Monte-Carlo simulations. HIGHLIGHTS: Promising preclinical studies for AD drugs have not translated to clinical success. Systematic assessment of AD drug candidates may increase clinical translatability. We describe a well-defined framework for compound selection with clear selection metrics.

PMID:37157089 | DOI:10.1002/alz.13108