Animal systems in the development of treatments for Alzheimer's disease: challenges, methods, and implications.

Substantial resources and effort have been invested into the development of therapeutic agents for Alzheimer's disease (AD) with mixed and limited success. Research into the etiology of AD with animal models mimicking aspects of the disorder has substantially contributed to the advancement of potential therapies. Although these models have shown utility in testing novel therapeutic candidates, large variability still exists in terms of methodology and how the models are utilized. No model has yet predicted a successful disease-modifying therapy for AD. This report reviews several of the widely accepted transgenic and nontransgenic animal models of AD, highlighting the pathological and behavioral characteristics of each. Methodological considerations for conducting preclinical animal research are discussed, such as which behavioral tasks and histological markers may be associated with the greatest insight into therapeutic benefit. An overview of previous and current therapeutic interventions being investigated in AD models is presented, with an emphasis on factors that may have contributed to failure in past clinical trials. Finally, we propose a multitiered approach for investigating candidate therapies for AD that may reduce the likelihood of inappropriate conclusions from models and failed trials in humans.