Elan Pharmaceuticals developed the first active Aβ vaccine to be tested in Alzheimer’s disease (AD) patients; it halted when 6 percent of treated patients developed meningoencephalitis (see ARF related news story). AN1792 used full-length Aβ42 together with an adjuvant that activated proinflammatory T helper cells, white blood cells that release inflammatory cytokines. “Our vaccine attempts to do the opposite,” said Lemere. It uses a short form of the Aβ peptide, invisible to T cells and recognizable only by B cells, along with an adjuvant that activates the Th2 pathway, in which T cells release anti-inflammatory cytokines.

To create MER5101, co-first authors Bin Liu and Jeffrey Frost and colleagues attached multiple copies of the Aβ1-15 peptide to diphtheria toxoid (DT). DT stimulated T cells and tricked them into presenting the Aβ antigen to B cells, which then produced antibodies against the Aβ fragment. Because T cells only recognize amino acids 16-42 of Aβ, they mounted no inflammatory response against the antigen. The researchers formulated their vaccine in Mercia’s MAS-1, which has generated robust antibody responses to endogenous antigens for elderly people and immune-compromised adults with cancer. As people age, their immune responses weaken, meaning that developing active vaccines in the elderly can be difficult.

Five times over four months, the scientists injected their vaccine under the skin of 10-month-old wild-type or APPswe/PS1ΔE9 transgenic mice, which develop Aβ deposits and cognitive impairments around six months of age. Immunized mice produced mostly IgG1 and IgG2b anti-Aβ antibodies—typically a result of anti-inflammatory Th2 signaling—with much lower levels of proinflammatory IgG2a antibodies. After these four months, vaccinated transgenic mice had 38, 29, and 42 percent less total Aβ in the hippocampus, frontal cortex, and cerebellum, respectively, than untreated controls. Vaccination also reduced thioflavin S-positive fibrillar plaques by about half in those same regions.

Microglia were less active in the hippocampus of immunized mice, which sported more synapses and fewer neuritic plaques than did untreated transgenics. Immunized transgenic mice also performed better in learning and memory tasks than controls did. They froze more in a contextual fear-conditioning task, and performed as well as wild-type mice on the Morris water maze. None of the mice showed microhemorrhages, which have become a concern in human Aβ immunotherapy (see ARF related news story); vaccination left vascular Aβ levels unchanged.

If a therapeutic antibody removes too much Aβ too quickly, it may lead to vasogenic edema (VE), as in the case of the higher doses of bapineuzumab used, noted Michael Agadjanyan, University of California, Irvine (see ARF related news story). The immunohistochemical data in the MER5101 paper suggest that the antibody binds weakly to fibrillar Aβ compared to the monomeric peptide, which could reduce VE risk, he added.

Lemere and colleagues are uncertain how MER5101 cleared plaques. A peripheral sink mechanism may be at work, they wrote. Plasma levels of Aβ-bound antibody rose and stayed high toward the end of the study, suggesting antibodies had pulled Aβ into the periphery. However, since insoluble and fibrillar brain Aβ also declined in vaccinated mice, some antibodies may have crossed the blood-brain barrier, bound existing plaques, and led to microglial phagocytosis, wrote the authors.

Agadjanyan is unconvinced a Th2 adjuvant is needed, arguing that any vaccine designed with a small, N-terminal piece of Aβ will avoid an inflammatory autoreactive T cell response. For instance, no such adverse events in people have been reported for ACC-001, which combines Aβ1-7 with DT and uses a strong Th1-type adjuvant, he told Alzforum (see ARF related news story). Agadjanyan recently reported on a tetanus toxin-based vaccine (see ARF related news story). Lemere believes that the active anti-inflammatory response from a Th2 adjuvant makes it advantageous, as many scientists suspect that inflammation plays a deleterious role in AD. “We believe that our two-pronged approach—combining antibody production and anti-inflammatory benefits from regulatory T cells—may be beneficial and separates our vaccine from others,” Lemere told Alzforum.

This vaccine will likely have a better safety profile than AN1792, said Dave Morgan, University of South Florida, Tampa. Morgan doubts that an excessive T cell response was the only problem in that instance. Other factors, such as microhemorrhage, could have played a part. Though Lemere and colleagues saw no microhemorrhages in their 14-month-old treated mice, Morgan and coworkers previously reported that only older APP/PS1 mice exhibit this problem (see Li et al., 2012). Lemere plans to test her vaccine in older animals and to incorporate other peptides, such as tau or pyroglutamate-3 Aβ into the MER platform for combination therapy. Mercia is looking for a pharmaceutical partner to help test the vaccine in humans, Lemere said.—Gwyneth Dickey Zakaib.

Reference:
Liu B, Frost JL, Sun J, Fu H, Grimes S, Blackburn P, Lemere CA. MER5101, a Novel Aβ1-15:DT Conjugate Vaccine, Generates a Robust Anti-Aβ Antibody Response and Attenuates Aβ Pathology and Cognitive Deficits in APPswe/PS1ΔE9 Transgenic Mice. J Neurosci. 2013 Apr 17;33(16):7027-37. Abstract