First off it was obvious from the Elan trials that suddenly making a protein abundant in plaques in the brain (amyloid beta) into an antigen would lead to inflammation.... what were they thinking!

Immunotherapy has been successfully used in multiple mouse models of AD, including peripheral active and passive immuunisation. But here is the question: Is removing extracellular amyloid plaques in human AD going to cure the disease?? The probable answer is no, more and more researchers are showing that plaques are an end point - a protective state of amyloid that traps free floating "harmfull" amyloid into a dense core where it cant do any harm.

The harmfull effects of amyloid are being shown to be mediated by the soluble and oligomeric species - that is a single amyloid peptide or a bunch stuck together, usually with a mass of less than 100kda. So far we dont know if immunotherapy in humans will affect these harmful "amyloids" or not. The post mortem results from the Elan trials were pesimestic at best - patients who recieved the injections had reduced amyloid plaque burden, but cognitively, at best (and this is from the company line) did not cognitively deteriate as fast as without the antibodies.

Either way I'll put my money on a nice BACE inhibitor. Forget about the gamma-secretase ones, thats one complex you really dont want to be messing with!