Initial Results Reported from a Phase 1 Safety Trial of a Tau Vaccine

Alzheimer's disease is both an amyloidosis and a tauopathy; its symptoms are produced by some combination of the presence of solid deposits of misfolded amyloid-β and tau protein, though there is much debate over which is more important and how they interact with one another and brain cells in order to produce pathology. Effective treatment will probably involve removing both amyloid-β and tau aggregates from brain tissue and cerebrospinal fluid. So far the best class of approach, and the one with the most funding behind it at the present, is immunotherapy, engineering the immune system to attack and remove the unwanted waste. Even that has proven to be much harder than expected, however, and the field is littered with failed trials and promising implementations that did not translate from animal studies to human biochemistry. Only recently have human trials produced meaningful results in amyloid clearance, and earnest efforts to remove tau from the brain started later and have less funding. Still, there is progress towards immunotherapies that can clear tau, as noted here:

So far, many of the antibody drugs proposed to treat Alzheimer's disease target only the amyloid plaques. Despite the latest clinical trial that is hailed as our best chance in the quest for treating Alzheimer's, all later phase trials have failed with many causing severe side effects in the patients, such as abnormal accumulation of fluid and inflammation in the brain. One of the reasons for side effects, many speculate, is due to the antibody directing a reaction towards normal amyloid present in blood vessels or simply releasing beta-amyloid caught in the vessel wall.

The authors of the study have developed a vaccine that stimulates the production of an antibody that specifically targets pathological tau, discovering its "Achilles' heel". It is able to do this because healthy tau undergoes a series of changes to its structure forming a new region that the antibody attacks. This new region (the "Achilles' heel"), while not present in healthy tau, is present in diseased tau early on. Therefore, the antibody tackles all the different varieties of pathological tau. In addition to this important specificity, the antibody is coupled to a carrier molecule that generates a considerable immune response with the added benefit that it is not present in humans, thus avoiding the development of an immune reaction towards the body itself.

Side effects have included a local reaction at the site of injection. This skin reaction is thought to occur due to the aluminum hydroxide, an adjuvant used in vaccines to enhance the body's own antibody production. No other serious secondary effects were directly related to the vaccine. Overall, the safety of the drug and its ability to elicit an immune response were remarkable. While many trials against Alzheimer's disease stubbornly continue to target amyloid, our study dares to attack the disease from another standpoint. This is the first active vaccination to harness the body's ability to produce antibodies against pathological tau. Even though this study is only a phase 1 trial, its success so far gives the authors confidence that it may be the answer they are looking for to halt the progress of this devastating disease.



Brain inflammation appears to be a problem in the use of IgG antibodies against amyloid beta, so I don't know why these investigators are so sanguine about using IgG antibodies against Tau? I read somewhere that groups are now trying to engineer IgG antibodies with defective FC regions in order to avoid some inflammation.

Also there is the Paul Sudhir paper that Michael Rae mentioned in previous comments were IgM catabolic antibodies are used to remove AB which seems like a much superior approach:

Maybe there could be a fundraiser next year to try and get a study going for IgM antibodies against Tau aggregates? Or perhaps a study to find IgM antibodies against glucosepane?

Posted by: Jim at December 13th, 2016 8:52 AM

IgG antibodies are standard antibodies of the adaptive immune system, and are the normal thing used to recognize peptide sequences/structures; it's just the obvious thing for which to reach. Really, Sudhir Paul is the only scientist working on IgM antibodies. You will not be surprised to learn that I'm personally more enthusiastic for that approach, but it's not at all an obvious or well-trod path.

I'm speculating, but it may also be more difficult to work up candidates for tau than Abeta: Dr. Paul started off using native IgMs purified from human plasma, selecting for those with the highest affinity for beta-amyloid, and while there are phosphorylated tau species in the CSF, I don't believe there are any in circulation (let alone more directly pathologically important species), so it's possible that no suitable IgM antibodies targeting aberrant tau in the circulation.

From their main animal study of AADvac1 vaccine, they don't seem to have been worried about cerebrovascular side-effects: they don't mention them, and don't look either in the brain vasulature proper or at the BBB. It's possible of course that they have unpublished data or some reason to strongly expect no problem. There isn't the equivalent of cerebral amyloid angiopathy in malformed tau, so it doesn't strike me as crazy to think that there may not be an equivalent for antibody-complexed tau either.

Aberrant tau species range only up to 70 kDa and most are down near 30-40 kDa, whereas some Abeta oligomers go up to >100 kDa, so there might be less of a problem of antibody-target complexes getting "stuck" at the BBB; that said, they do claim that the antibody captures early and mature neurofibrillary tangles in their mice, and these are big structures. And, counter to this line of thinking, solenezumab captures monomers, and yet it causes cerebrovascular oedema anyway by mobilizing the stuff and not getting across the BBB.

For what it's worth, however, they do report that over the course of 24 weeks (12 weeks treated, 12 weeks open-label extension) "No cases of meningoencephalitis or vasogenic oedema occurred after administration [and just] One patient with pre-existing microhaemorrhages had newly occurring microhaemorrhages" in 24 mild-to-moderate Alzheimer's disease patients, ages 50-85.

They seem more worried about Th1-mediated inflammatory response like the one that sunk AN1791, the original (active) Abeta vaccine. They found no evidence of such in their animal studies, but rather a consciously-generated T-helper cell-based Th2 response.

Looking again at their animal studies, their target is a motif they find in many pathological tau species, but they actually seem to be specifically strategizing that they will hit tau species in the very early stages of tau oligomerisation, preceding tau hyperphosphorylation, and preventing it from forming more advanced structures. Most of their data seems consistent with this, and they only directly document slowing the progression of gross neuropathology rather than actual clearance, so maybe there is no capturing going on or only capturing of small and easily-trafficked species - or maybe they're sufficiently labile as to be degraded once bound and prevented from further phosphorylating or oligomerizing.

We'll just have to wait for more animal and/or human work. Meanwhile, we should celebrate that vaccines targeting aberrant tau are now being tested, and that one has shown preliminary safety and strong immunogenicity. We need these vaccines to move forward, and happily several are, with distinct targets and mechanisms of action, so we should be able to get an idea on how promising each approach is.

Posted by: Michael at December 13th, 2016 7:04 PM

Re targeting intra-cellular Tau. I read this piece on GenEngNews yesterday about using mRNA for enzyme replacement (for what I assume are enzymes that are not easily taken up by cells from the bloodstream).

Perhaps it would be worthwhile searching for a bacterial enzyme that breaks up Tau, then using this technology to deliver it. Of course given the massive costs to society of Alzheimer's disease it is nuts that this isn't already being investigated by multiple teams around the world without the SENS foundation and others having to push things along.

Posted by: Jim at December 14th, 2016 8:48 AM

Post a comment; thoughtful, considered opinions are valued. New comments can be edited for a few minutes following submission. Comments incorporating ad hominem attacks, advertising, and other forms of inappropriate behavior are likely to be deleted.

Note that there is a comment feed for those who like to keep up with conversations.