Abstract
Animal models of human brain disorders permit to explore disease mechanisms and to test potential therapies. However therapeutic molecules derived from animal models often translate poorly to the clinic. While human data may be more relevant, experiments on patients are constrained and living tissue is unavailable for many disorders. Here we compare work on animal models and on human tissue for three epileptic syndromes where human tissue is excised therapeutically. (1) acquired temporal lobe epilepsies, (2) inherited epilepsies associated with cortical malformations and (3) peritumoural epilepsies. Animal models rest on assumed equivalencies between human brains and brains of mice, the most frequently used model animal. We ask how differences between mouse and human brains could influence models. General principles and compromises in model construction and validation are examined for a range of neurological diseases. Models may be judged on how well they predict novel therapeutic molecules or new mechanisms. The efficacy and safety of new molecules are evaluated in clinical trials. We judge new mechanisms by comparing data from work on animal models with those data from work on patient tissue. In conclusion we stress the need to cross-verify findings from animal models and from living human tissue to avoid the assumption that mechanisms are identical.
FEB