See other treatment stories 1, 2, 3, and 5.

A novel therapeutic approach captured attention at the second Alzheimer’s Association International Conference on Prevention, which wound down today in Washington, D.C. Lauren Costantini of the biotechnology company Accera in Broomfield, Colorado, presented data of a short phase 2b trial of a milkshake to be taken with breakfast. The active ingredient in this beverage is a compound that the liver converts to ketone bodies. These then enter the brain and provide an alternative source of energy for neurons that no longer metabolize glucose effectively.

That cerebral glucose metabolism wanes years before AD is diagnosed is fairly well established scientifically. The question here is whether an alternative, downstream entry point into the metabolic cascade that eventually generates the cellular fuel ATP can restore the neuron’s metabolism enough to noticeably help people with AD. This is a treatment that is not by itself specific to AD. Rather, it would prop up ailing mitochondria and tackle the energy deficit that plagues the brain in several neurodegenerative diseases. If all goes well, it could become part of a combination therapy for AD, Costantini said.

A previous double-blind, placebo-controlled study in an aged dog model of dementia indicated that the compound, called AC-1202 and trademarked Ketasyn, made the old dogs more active and moved some readouts of mitochondrial efficiency and oxidative damage in the desired direction, Costantini said. A subsequent single-dose phase 2a study in 20 people with mild to moderate AD showed that ketone body concentration in people’s blood rose; it also showed an uptick on the ADAS-cog scale in people who do not carry the ApoE4 allele.

At the conference yesterday, Costantini described the latest phase 2b study of 152 people aged 50 and older who had mild to moderate AD and also took a cholinesterase inhibitor, conducted at 25 sites in the U.S. They were randomized to drug or placebo, took it for 3 months, had it washed out for 2 weeks, and then continued in an open-label extension for another 6 months.

Side effects were minor, mostly diarrhea for the first month until the digestive system had adjusted to handling the elevated number of ketone bodies (i.e., fats) coming out of the liver.

When analyzing all patients randomized into the trial together, the treatment showed a small difference in the ADAS-cog endpoint though that was not statistically significant. The most interesting finding was a difference between patients depending on which allele of the ApoE risk gene they had inherited. People with the 2 or 3 alleles, who make up half of all AD patients nationwide, benefited from the milkshake cognitively and on the overall clinical assessments, while people who carry the 4 allele did not. This is the second time this pharmacogenomic effect has turned up in an AD application of what is primarily a metabolic drug. The diabetes drug rosiglitazone also appears to help people with ApoE2 or 3, but not 4 (Risner et al., 2006; also see ARF related conference news on insulin and AD), and the pharmaceutical company GSK is currently testing that in large phase 3 trials.

If such differences become firmly established, and the requisite drugs approved for AD, doctors may begin recommending ApoE genotyping for some patients, as that knowledge would help them assign a therapy that is likely to help and at the same time avoid exposing ApoE4 carriers to unnecessary risk of side effects. Right now, doctors have no such way of distinguishing which drug is more likely to work in which patient. The Alzforum is hosting a Webinar on the topic of ApoE genotyping this Friday, June 15.

Lastly, on a basic science note, the mechanism of this drug brings up an intriguing similarity with what the ketogenic diet does for severe cases of pediatric epilepsy. There, kids are fed a high-fat diet. They are asked to eat large helpings of mayonnaise, for example. The children understandably dislike the diet, but it controls their seizures better than drugs. Mechanistic research by Gary Yellen at Harvard Medical School has begun describing how it works. Reportedly, the ketone bodies enhance a normal gating function of normal potassium channels, which serves to limit the rate of spontaneous firing (Ma et al., 2007; see also Gasior et al., 2006). AD and epilepsy are very different diseases, to be sure. Yet some AD researchers are beginning to develop an interest in epileptiform activity in AD mouse models, and it centers around the regulation of synaptic transmission. People with AD, especially the early onset familial forms, have been reported to have seizures. This is still a poorly understood area of science. For example, there is some data linking epilepsy drugs to increased risk for developing dementia (Carter et al., 2007). It’s unclear whether in AD, one would want to dampen dysregulated synaptic transmission or boost transmission of dysfunctional synapses, and how that could be done with regional specificity.—Gabrielle Strobel.


  1. This piece of work by Accera scientists is very exciting and clearly offers an intriguing new angle to ApoE genotype-driven treatments. In this context it is interesting to recall the original 1995 observation that tacrine-treated cases exhibited a much stronger drug response in the non-E4 cases as opposed to ApoE4 carriers (1). A few years ago, probucol, a first-generation cholesterol-lowering agent used to treat hypercholesterolemia in humans, was also found to affect the disease progression and CSF β amyloid levels in an ApoE genotype-dependent manner (2), the best responders belonging to the ApoE4-negative group.

    These results, combined with many recent industry-sponsored pharmacogenomic analyses in MCI and AD (3,4), clearly support a more proactive integration of pharmacogenomic markers such as ApoE, butyrylcholinesterase, and paraoxonase in the design of the next generation of clinical drug trials.


    . Apolipoprotein E4 allele as a predictor of cholinergic deficits and treatment outcome in Alzheimer disease. Proc Natl Acad Sci U S A. 1995 Dec 19;92(26):12260-4. PubMed.

    . Apolipoprotein E and cholesterol metabolism in the pathogenesis and treatment of Alzheimer's disease. Trends Mol Med. 2003 Mar;9(3):94-101. PubMed.

    . Vitamin E and donepezil for the treatment of mild cognitive impairment. N Engl J Med. 2005 Jun 9;352(23):2379-88. Epub 2005 Apr 13 PubMed.

    . Effect of butyrylcholinesterase genotype on the response to rivastigmine or donepezil in younger patients with Alzheimer's disease. Pharmacogenet Genomics. 2006 Nov;16(11):771-4. PubMed.

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. Washington: Alzhemed Non-story Yields Spotlight to Phase 2 Treatments
  3. Washington: γ-secretase Inhibitor Survived Phase 2, Moving to 3
  4. Washington: Functional Improvement From Stopped Vaccine
  5. Madrid: Highs and Lows of The Insulin Connection

Webinar Citations

  1. Susceptibility Testing and Risk Assessment in Alzheimer Disease

Paper Citations

  1. . Efficacy of rosiglitazone in a genetically defined population with mild-to-moderate Alzheimer's disease. Pharmacogenomics J. 2006 Jul-Aug;6(4):246-54. PubMed.
  2. . Ketogenic diet metabolites reduce firing in central neurons by opening K(ATP) channels. J Neurosci. 2007 Apr 4;27(14):3618-25. PubMed.
  3. . Neuroprotective and disease-modifying effects of the ketogenic diet. Behav Pharmacol. 2006 Sep;17(5-6):431-9. PubMed.
  4. . Epilepsy and antiepileptic drug use in elderly people as risk factors for dementia. J Neurol Sci. 2007 Jan 31;252(2):169-72. PubMed.

Other Citations

  1. early onset familial forms

Further Reading

No Available Further Reading