Overview. One of the most exciting prospects in the treatment of Alzheimer’s disease is the development of a therapeutic immunization strategy that could prevent disease progression by clearing amyloid plaques from the brains of Alzheimer’s disease patients. This excitement, however, has been tempered by the discontinuation of the Phase Ha trial of Elan’s vaccine AN-1792 (Betabloc). Elan reported in March 2002 that the trial was terminated because 18 of the approximately 300 treated patients developed fatal meningoencephalitis.
Although the AN-1792 active immunization trial was suspended, many companies are still pursuing immunization. Two types of Aβ immunization strategies are being tested for the treatment of Alzheimer’s disease: “active” immunization, or vaccination, which involves triggering a patient’s immune response against an injected antigen, and “passive” immunization, which involves injecting an anti-Aβ antibody, generated ex vivo, into an Alzheimer’s disease patient. Elan and Wyeth are collaborating on a second-generation Alzheimer’s disease vaccine, ACC-001. Neurochem, Lundbeck/Pharmexa, VLP Biotech, Prana Biotechnology/Panvax, and Pevion Biotech are also working on similar active vaccination strategies.
Mindset Pharmaceuticals has inlicensed an active immunization strategy from researchers at the New York University School of Medicine. The researchers have synthesized and modified portions of the Aβ peptide in an effort to make them nontoxic. The vaccine was shown to inhibit Aβ deposition in mouse models of Alzheimer’s disease and to improve the cognitive performance of the mice in memory tests. The vaccine remains in discovery phase. Lastly, Acumen and Merck have entered a collaboration to design active and passive vaccines against small Aβ fibrils known as amyloid-derived diffusible ligands, or ADDLs, which are increasingly implicated as the neurotoxic species in Alzheimer’s disease. Antibodies against ADDLs have prevented and even reversed memory deficits in animal models of Alzheimer’s disease, an exciting effect if clinical trials bear these results out.
In the area of passive immunization strategies, Eli Lilly has developed an antibody, m266, to be used in passive immunization approaches. Elan and Wyeth are collaborating on strategies for amyloid plaque removal with the development of the humanized anti-amyloid antibody AAB-001. The companies filed an investigational new drug (IND) application for AAB-001 in August 2003, and Phase I trials began in June 2004. While this antibody may produce results similar to the m266 antibody developed by Lilly, it is also subject to the same potential side effects.
ACC-001. ACC-001 is an active-immunization antigen developed by Elan and Wyeth for the treatment of patients with mild to moderate Alzheimer’s disease. Unlike the AN-1792 vaccine, in which the antigen was the entire 42-amino acid Aβ fragment, ACC-001 is a peptide fragment of amyloid beta (). This strategy presents the advantage of generating fewer nontherapeutic antibodies (binding antibodies) against Aβ; indeed, binding antibodies increase the risk of developing inflammation.
The development of a refined vaccine using a portion of the Aβ peptide instead of its entirety may yield promising results. Although the exact Aβ portion used for ACC-001 has not been identified, it is most likely a portion of the N-terminus of Aβ, which has been shown to be more antigenic than the C-terminus of the peptide. E.M. Sigurdsson and colleagues at the University of Toronto characterized therapeutic antibodies in mouse models of Alzheimer’s disease by mapping the region of Aβ recognized by therapeutic antibodies (). Therapeutic antibodies recognized a region on the N-terminus of Aβ, between amino acids 4 and 10. These antibodies inhibited fibrillogenesis, disrupted preexisting Aβ fibers, and rescued spatial memory deficits observed in the mice. Inhibition of Aβ fibrillogenesis was not seen with antibodies recognizing different regions (epitopes) within the Aβ sequence. Importantly, the inhibition of fibrillogenesis was not associated with an inflammatory response, underscoring the fact that active vaccination approaches need not be associated with the side effects seen in the AN-1792 trials. While the region recognized by ACC-001 may not be the Aβ region between amino acids 4 and 10, these data suggest the potential for a safe immunotherapy approach using a portion of the Aβ peptide.
One important caveat is that the antigen used for immunotherapies in transgenic mouse models of Alzheimer’s disease is of human origin and is targeted against the human amyloid precursor protein transgene introduced into these mice. If an effective immune response clears human Aβ from transgenic mice, these mice still retain their murine amyloid precursor protein, which may fulfill important functions and prevent the development of serious side effects. These side effects may become apparent in human patients treated with the vaccine because these patients would have no compensatory Aβ species once their Aβ had been cleared.
M266 (Passive Immunization). Eli Lilly has developed an antibody, m266, against the central region of Aβ and plans to use the antibody in passive immunization in Alzheimer’s disease patients. M266 is in discovery phase, but preclinical trials demonstrate that m266 lowers the levels of amyloid plaque in the cortex and hippocampus of a mouse model for Alzheimer’s disease ().
M266 does not cross the blood-brain barrier, but a single injection of m266 causes an increase of Aβ in the blood in a matter of minutes. Because Aβ is produced almost exclusively in the brain, this indicates that the antibody acts by drawing Aβ out of the brain, a theory known as the “peripheral Aβ sink” hypothesis.
Long-term m266 treatment does not affect the plaque burden in the brains of mouse Alzheimer’s disease models, suggesting that the antibody works either by preventing the early deposition of Aβ peptide into plaques or by enhancing the peripheral clearance of the Aβ/m266 complex (). The antibody is capable of preventing Aβ fibril formation in vitro, however, an important result if the neurotoxic species in Alzheimer’s disease is Aβ oligomers instead of amyloid plaques (). Importantly, a single m266 injection in older mice also rapidly reduces the levels of error in a learning and memory task in 24 to 72 hours ().
Although side effects associated with monoclonal antibody therapy in multiple sclerosis patients (e.g., fatigue, chills, diaphoresis) are mild (), the potential for several side effects in a passive Alzheimer’s disease vaccine approach exist. Most troublesome is the possibility that the patient’s immune system will mount a reaction to the injected antibody itself. Some of these antibodies (neutralizing antibodies) may reduce the therapeutic efficacy of m266 if they recognize, and interact with, the site of the m266/Aβ interaction. In addition, cerebral hemorrhages have also been reported in the brains of mice treated with Aβ antibodies, although these results may be a function of the strain of mouse used in the study (). These side effects may be reduced by cleaving the monoclonal antibody and using its active, specific portion (the [Fab’]2 fragment) and removing the Fc portion, which activates the cellular immune response. Such a modified Aβ antibody has been shown to clear Aβ in a mouse model ().