Recent reports indicate that amyloid-beta (Abeta) vaccine-based therapy for Alzheimer's disease (AD) may be on the horizon. There are, however, concerns about the safety of this approach. Immunization with Abeta1-42 may not be appropriate in humans because it crosses the blood-brain barrier, can seed fibril formation, and is highly fibrillogenic. Abeta1-42 fibrils can in turn cause inflammation and neurotoxicity. This issue is of a particular concern in the elderly who often do not mount an adequate immune response to vaccines. Our findings show that vaccination with nonamyloidogenic/nontoxic Abeta derivative may be a safer therapeutic approach to impede the progression of Abeta-related histopathology in AD. Although the site of action of the anti-Abeta antibodies has been suggested to be within the brain, peripheral clearance of Abeta may have a greater role in reducing cerebral amyloid plaques in these animals and eventually in AD patients. Antibodies in general are predominantly found outside the central nervous system (CNS) and will, therefore, primarily clear systemic Abeta compared to brain Abeta. This disruption of the equilibrium between central and peripheral Abeta should then result in efflux of Abeta out of the brain, and subsequent removal of plaques. Abeta therapy can be targeted to the periphery, which may result in fewer CNS side effects, such as inflammation. Future Abeta derived vaccines should include T(h) epitopes, carriers and/or lipid moieties to enhance antibody production in the elderly, the population predominantly affected by AD.