The 8th Clinical Trials on Alzheimer’s Disease conference, held November 5-7 in Barcelona, Spain, featured results on at least four clinical trial debutants, i.e., Phase 1 entries. One evoked Queen Marie Antoinette’s acerbic observation, “There is nothing new except what has been forgotten.” For those who have forgotten tramiprosate, aka Alzhemed, aka Vivimind—well, it’s back, and the new version is better than ever if you believe its new sponsors. At CTAD, Anton Porsteinsson of the University of Rochester School of Medicine and Dentistry in New York told the story of how this failed medication is making its comeback.

Alzhemed’s active ingredient is homotaurine, aka 3-amino-1-propanesulfonic acid. Naturally found in seaweed, it is an analog of taurine, an amino acid ingredient of some brands of infant formula and energy drinks such as Red Bull. Alzhemed underwent a clinical development program in the early 2000s. When a North American Phase 3 trial missed both primary endpoints, a concurrent European Phase 3 study was stopped early. The compound was subsequently rebranded and sold as a nutraceutical, unregulated by the FDA (Jun 2007 conference newsNov 2007 news; Oct 2008 news). 

In 2013, the Lexington, Massachusetts-based startup company Alzheon picked up this thread. It licensed a prodrug version, ALZ-801, whose pharmacokinetic and -dynamic properties it claims will cause fewer gastrointestinal complaints, achieve more consistent drug levels in the blood, and last longer so that patients need take a pill only once, not twice, a day. Company scientists further claim that the rubble of the Alzhemed program contains promising signals of efficacy in certain subgroups. Those signals, combined with some of Alzhemed’s safety data the FDA may accept, could speed up development of this approach the second time around, Porsteinsson said. Porsteinsson serves on Alzheon’s scientific advisory board.

Porsteinsson presented a post hoc analysis of the ApoE4 homozygous participants of the two Alzhemed Phase 3 studies. Together, the two trials contained 252 ApoE4/4 patients with mild to moderate AD. Unlike the whole study population, those among this subgroup who received the high dose had a statistically significant treatment benefit on the ADAScog of 2.85 and 4.56 points at 52 and 78 weeks, as well as a trend on the CDR-sb, Porsteinsson said.

In the meantime, Alzheon on November 6 issued a press release about starting two Phase 1b trials. One is a dose-ranging study assessing ALZ-801 tolerability and pharmacokinetics in healthy elderly volunteers, the other a bioequivalence study to assess pharmacokinetics and food effects of a tablet formulation. Details were not disclosed, and these trials are not listed in At CTAD, Porsteinsson said the next step would be a Phase 2/3 study comparing one dose of Alz-801 to placebo in 200 patients each with mild to moderate Alzheimer’s disease who are positive for ApoE4. The trial will likely last for up to 78 weeks, Porsteinsson said, and the ADAScog and CDR-sb will serve as co-primary outcomes.

Other scientists at CTAD noted that homotaurine is a GABA A receptor agonist (Caltagirone et al., 2012), and that tramiprosate’s amyloid-reducing mechanism was not rigorously validated in a clinical trial, nor target engagement conclusively shown in humans.

With the ALZ-801 program intending to skip Phase 2, it’s arguably farther along than two other Phase 1 candidates presenting at CTAD, both of which are transitioning to Phase 2. One is AAD-vac1, an active vaccine designed to prompt the formation of antibodies against a pathological conformation of tau protein. At CTAD, Matej Ondrus of AXON Neuroscience in Bratislava, Slovakia, presented results of the first Phase 1 trial. It enrolled 30 people with mild to moderate Alzheimer’s and randomized them to one of two doses of the vaccine or placebo, injected under the skin once a month, for a three-month blinded period followed by three months of open-label treatment for everyone. On safety, Reinhold Schmidt of the University of Graz, Austria, had previously reported at AAIC that the study had two withdrawals because of adverse events, one of which might have been due to the study medication. Besides that, the vaccine appeared safe and well-tolerated.

At CTAD, Ondrus added that 29 of the 30 participants responded to the antigen by making antibodies against what Axon calls peptide 108 with titers that rose with each dose after the second injection. Some commentators considered the presented titers high for an aging population. Vaccines against Aβ have struggled to elicit both high response rates and high titers. AAD-vac1 uses the adjuvant aluminum hydroxide, aka Alhydrogel.

Ondrus showed ADAS-cog scores, noting that their variability in this small sample precluded a statistical analysis but that the participants appeared to stay stable for the six-month course of the trial. Data correlating cognitive response to titers in individual patients were not shown. In response to a question about CSF biomarkers, Ondrus noted that lumbar puncture was optional in this trial. He said four patients completed it, and the data were still being assessed. The trial concluded in March 2015. A Phase 2 trial is set to enroll its first patient this December, Ondrus noted.

Another Phase 1 tau immunotherapy, Roche’s RG7345, appears to have bitten the dust. This monoclonal antibody directed against the tau phosphoepitope pS422 was in Phase 1, but in October 2015, Roche removed it from its active development pipeline.

A repurposed drug just advanced from Phase 1. AZD0530, aka saracatinib, is a tyrosine kinase inhibitor originally developed to treat cancer. It is now being used as a way to test fyn kinase as a therapeutic target in Alzheimer’s disease. At CTAD, Christopher van Dyck of Yale University School of Medicine in New Haven, Connecticut, told the audience that doing so is tricky. “This drug does not appear to have a wide therapeutic index. The effective brain levels that we are trying to achieve are close to those where toxicity starts,” van Dyck said. This means that drug levels need to be monitored to keep trial participants safe.

In a Phase 1b dose-finding trial, 50 mg/day of saracatinib fell short of achieving adequate CSF levels of the drug, 100 mg achieved them in half of the participants, and 125 caused tolerability or safety problems in some participants. Diarrhea, headache, nausea, and fatigue were common adverse events, and one person was hospitalized with congestive heart failure and pneumonia, which were considered possibly related to study drug, van Dyck said. To optimize dosing in individual participants, the scientists are monitoring blood levels, as the drug’s plasma concentration accurately signal its CSF concentration, van Dyck said.

The Phase 2 study is being coordinated by the Alzheimer’s Therapy Research Institute at the University of Southern California. For this 152-patient, 23-center trial of a one-year course of saracatinib, patients randomized to drug start on 100 mg/day and increase to 125 mg only if their plasma drug level is deemed to be insufficient, van Dyck said. Because this is a proof-of-concept study, not a registration trial, van Dyck opted for a biomarker that reflects the drug’s mechanism of action. Saracatinib’s effect on fyn kinase is thought to affect synaptic activity, hence FDG-PET appears to be a logical way to measure if it is biologically active in the brain of Alzheimer’s patients, van Dyck said.

a id="earliest" name="earliest">The earliest results in a new Phase 1 program came from MEDI1814Amanda Dudley of AstraZeneca said that the first clinical trial for this antibody is beginning to show target engagement in the central nervous system. Dudley presented an interim analysis of this ongoing study. MEDI1814 is the latest in a long list of antibodies directed against Aβ42. Its unique wrinkle is that it arose from a different approach toward reducing ARIA than its competitor crenezumab. That antibody, now entering Phase 3, reduced microglial activation by engineering the Aβ recognition region onto a IgG4 antibody isotype instead of the more common IgG1. MEDI1814 is an IgG1; to reduce effector function, its FC tail carries mutations that interfere with binding to microglial receptors, Dudley said. This antibody came about when AstraZeneca bought MEDIMMUNE, a company better known for its influenza vaccines and focus on cancer, inflammatory and autoimmune diseases.

The first trial is an ascending study, in which a group of people with mild to moderate Alzheimer’s get a single or multiple injections. Once that proves safe, a second group gets a higher dose, etc. The sites monitor for adverse events, conduct MRIs at baseline and after five weeks to look for ARIA, and measure the antibody’s behavior in blood and CSF.

Within a dose range of 25 to 300 mgs, data are available thus far up for doses up to 40 mg in the first 21 of 121 planned patients, Dudley told the audience. No safety signals have cropped up yet. The antibody’s rise and fall in blood follows what was modeled previously, Dudley said. As expected, total plasma Aβ42 goes up as it binds the antibody and assumes the antibody’s half-life, which is longer than that of unbound Aβ42. Most importantly, in the CSF, the concentration of free Aβ42 drops, along with a rise of total Aβ42. This happens in a dose-dependent manner, Dudley said, noting that thus far in the trial, she has seen free CSF Aβ42 go down by 60 percent. Aβ40 does not change. Conducted at eight centers, this trial will continue for the better part of next year, Dudley said.—Gabrielle Strobel


No Available Comments

Make a Comment

To make a comment you must login or register.


News Citations

  1. Washington: Site-to-Site Differences Delay Result of Alzhemed Trial
  2. Philadelphia: European Trial of Alzhemed Ends, Marketing Morphs to Supplement
  3. Experts Slam Marketing of Tramiprosate (Alzhemed) as Nutraceutical

Therapeutics Citations

  1. AADvac1
  2. RG7345
  3. Saracatinib
  4. MEDI1814

Paper Citations


External Citations

  1. press release

Further Reading

No Available Further Reading