29 August 2008. While the glare of the media focuses on the trials and tribulations of the next potential billion-dollar pharmaceutical treatment of AD, other investigators have been quietly cooking up alternative approaches. One of those was on the menu at the International Conference for Alzheimer’s Disease (ICAD), held 26-31 July in Chicago. Philip Scheltens of Vrije University Medical Center in Amsterdam, the Netherlands, is well established in the AD field for his extensive brain imaging research in Alzheimer disease and related dementias. At ICAD, however, Scheltens presented data for a proof-of-concept human trial of a multi-nutrient drink. The drink is a product of Numico Research in Wageningen, The Netherlands, which in July 2007 was bought by Danone (Dannon to Americans). The drink is a defined mixture of ingredients said to be needed for membrane synthesis and synapse formation. According to Scheltens, it yielded a promising signal in a controlled 12-week trial of 212 people with early AD.

The idea behind this approach is that the synapse loss that characterizes AD might slow down if the ailing brain had an ample supply of synaptic membrane constituents, such as phosphatides, more readily available. These membrane phosphatides, in turn, are synthesized from molecular ingredients that were shown in previous research to influence learning and memory in various animal models. Much of the prior research over the past decade that has led up to the present trial comes from the laboratory of Richard Wurtman at the Massachusetts Institute for Technology (see, for example, Cansev et al., 2008; Wurtman et al., 2006; Wang et al., 2005; Holguin et al., 2008).

Scheltens and colleagues tested a drink formulated to contain ingredients in six functional categories previously studied by Wurtman’s and other groups: uridine-5’-monophosphate (UMP), omega-3 fatty acids (see also below for update on DHA trial), choline, phospholipids, B vitamins, and antioxidants. This mixture is trade-named Fortasyn^TM Connect, and the drink goes by Souvenaid^TM.

Conducted at 28 sites in The Netherlands, Belgium, Germany, and the U.S., the trial was randomized, double-blind, and controlled with a second drink that tasted the same as Souvenaid. “It tastes good; people liked it,” Scheltens said. One hundred six patients who were recently diagnosed with probable AD as per NINCDS-ADRDA criteria drank Souvenaid, and 106 placebo, once a day for 12 weeks. After that, they could choose to continue the blinded study; 160 patients finished the full 24 weeks. At ICAD, Scheltens presented the 12-week data. Primary outcome measures were a delayed verbal memory task (WMS-r) and ADAS-Cog-13, an extended ADAS-Cog battery. Secondary outcomes included functional, quality of life and behavioral measures, as well as plasma levels of nutrients and homocysteine, and safety measures. This initial trial included no CSF measures to show brain penetration and exposure data, nor brain imaging. The patients in this trial did not take acetylcholinesterase inhibitors or Namenda. The trial is not on clinicaltrials.gov but was listed with the Dutch Trial Register.

The groups were well matched, Scheltens reported, and Souvenaid produced no worrisome safety signals. A number of side effects were more common in the Souvenaid group, though none of the differences reached statistical significance. Some 18 percent in both control and Souvenaid groups had diarrhea, constipation, or nausea. Upwards of 92 percent of patients complied with the regimen. In plasma at 12 weeks, Souvenaid increased the concentration of the omega-3 fatty acid DHA (see more below), and reduced that of homocysteine, a metabolite thought to hasten cognitive decline.

Also at 12 weeks, Souvenaid improved performance of the WMS-r memory test. This effect was statistically significant, and the effect size was larger in a pre-specified subgroup of people who had an MMSE of greater than 23 at study onset, i.e., very mild AD, than in the more impaired group. The ADAS-Cog curves overlapped, showing no effect in the treatment group and no decline in the control group. A pre-specified analysis looking at baseline MMSE value as a predictor showed a greater numerical effect in the milder cases, but this did not reach statistical significance.

Analysis of the secondary outcomes as well as of the 24-week data is still to come, Scheltens said. “We know Souvenaid has a quick effect,” he added. “Whether it lasts past 12 weeks we do not know yet.” For now, he emphasized that Souvenaid seems to be safe and well tolerated, and has shown a signal of efficacy that serves as a proof of concept to support further clinical study of this product.

Fortasyn^TM is not to be mixed up with Ketasyn^TM, another breakfast shake drink meant to shore up starving neurons in AD, but containing different ingredients (see ARF related conference story).

Also at ICAD, there were tidbits of news on one component of the Souvenaid mixture, i.e., the polyunsaturated omega-3 fatty acids. One of those is docosahexaenoic acid (DHA), which in its own right has a growing body of epidemiological and experimental evidence suggesting it may protect against age-related cognitive decline (see, for example, Calon et al., 2004; Freund-Levy et al., 2006; Freund-Levy et al., 2007). Several presentations from different groups converged around the theme that DHA, and perhaps other nutrient-based approaches, show the most robust efficacy signal early on, that is at the MCI or very early AD stage, but help little by the time a person has moderate AD.

On DHA, the Alzheimer’s Disease Cooperative Study reported that enrollment of its DHA trial has been completed in 10 months. This trial tests a recombinant version of DHA generated in microscopic algae by Martek Biosciences, which is widely used as an ingredient of infant formula. Previously, this form of DHA had been shown to reduce accumulation of both amyloid and tau pathology in triple-transgenic mice, among other models (see Green et al., 2007).—Gabrielle Strobel.

This concludes a three-part series. See also Part 1 and Part 2.