Recent clinical trial results present a mixed bag of good and bad news. On the good news front, a regimen of high-dose B vitamins slowed the rate of brain shrinkage in people with mild cognitive impairment, suggesting these vitamins could have value in treating neurodegeneration. The VITACOG trial was led by David Smith and Helga Refsum at the University of Oxford, U.K., and the report appeared in the September 8 PloS One online. On the bad news front, a Phase 3 trial of the insulin-sensitizing drug rosiglitazone failed to find any cognitive benefit, scuttling development of this once-promising therapeutic. The results are reported in the August 21 Dementia and Geriatric Cognitive Disorders online by a research group led by Michael Gold at the pharmaceutical company GlaxoSmithKline, in Research Triangle Park, North Carolina, and were previously announced at ICAD 2009 in Vienna, Austria (see ARF related news story). That finding raises doubts about further testing of other drugs in this thiazolidinedione class, such as pioglitazone, which recently passed a Phase 1 safety trial in people with cognitive impairment. Additionally, drug regulators in the U.S. and Europe announced yesterday that rosiglitazone (trade name Avandia) would no longer be widely available as a diabetes medication due to increased risk of heart disease in those taking the drug (see New York Times story). In Europe, approval has been withdrawn, while in the U.S., it will only be available as a last resort.

The rationale behind the VITACOG trial was that high doses of three B vitamins—folate, B6, and B12—can lower plasma levels of the amino acid homocysteine. Homocysteine is a byproduct of cellular metabolism that is known to be neurotoxic in vitro, and in animal models has been shown to increase the formation of phosphorylated tau. Numerous cross-sectional and prospective studies have shown that elevated plasma homocysteine is associated with impaired cognition and a higher risk of dementia (for one example, see results from the Framingham Study, Seshadri et al., 2002). High homocysteine has also been correlated with shrinking brain volume in several papers.

A handful of previous studies have looked at the cognitive effect of lowering homocysteine, with mixed results. A two-year clinical trial found no cognitive improvement from B vitamin supplements in healthy elderly people with raised homocysteine (see McMahon et al., 2006), while a similar trial on healthy elderly volunteers found that folate supplements improved cognition (see Durga et al., 2007). However, B vitamin supplementation for more than five years did not delay cognitive decline in a population of cognitively normal elderly women, except in those with low baseline dietary intake of B vitamins (see Kang et al., 2008). The Alzheimer’s Disease Cooperative Study conducted a trial of high-dose B vitamins on 340 participants with mild to moderate AD, and found treatment did not slow cognitive decline (see ARF related news story on Aisen et al., 2008).

Refsum and colleagues took a different approach, enrolling 271 volunteers with mild cognitive impairment (MCI). About 50 percent of people with MCI convert to dementia within five years, said first author David Smith, so this is a population at high risk. Participants received either high-dose B vitamins (0.8 mg/day folic acid, 0.5 mg/day B12, 20 mg/day B6) or placebo for two years. The main outcome measure was the change in the rate of atrophy of the brain, as assessed by volumetric MRI scans. The normal aging brain shrinks at a rate of about 0.5 percent per year, Smith said, while the brains of people with MCI decline at twice that rate, and those with AD lose 2.5 to 3 percent of brain volume per year.

The authors found that B vitamin treatment decreased brain shrinkage by about one-third, from more than 1 percent per year in the placebo group to an average of 0.75 percent per year in treated volunteers. However, the effect depended on participants’ initial homocysteine levels. People with initial homocysteine levels below 9.5 μmol/L derived no benefit from B vitamins, while those with starting levels above 13 μmol/L saw a 50 percent reduction in their rate of atrophy. The trial was not powered to detect cognitive effects, Smith said, but nonetheless they are analyzing that data now and are seeing some trends, which they expect to publish. Smith presented preliminary data at the 2010 ICAD in Honolulu, Hawai’i, showing that volunteers with initially high homocysteine levels who received B vitamins had significantly slower decline in episodic memory than the placebo group.

One difference from previous B vitamin trials, Smith said, is that the VITACOG trial looked at people with MCI rather than healthy elderly or those with AD. By the time cognitive decline has progressed to dementia, Smith suggested, neuronal degeneration may be too far advanced for vitamins to slow it. Smith pointed out that in the ADCS trial of B vitamins, led by Paul Aisen at the University of California, San Diego, subgroup analysis showed that those with mild AD had some cognitive improvement, whereas those with moderate AD did not. Another feature of the VITACOG trial, Smith said, was that they measured brain atrophy rather than working from any hypothesis about the mechanisms of AD. “We believe atrophy is an important substrate of cognitive decline,” he said, adding that the results suggest that B vitamin treatment is modifying the underlying disease process. Smith and colleagues now plan to conduct a clinical trial of high-dose B vitamins on about 1,000 volunteers to see if treatment can reduce the conversion rate from MCI to AD.

“This is a key paper in the homocysteine story,” said Andrew McCaddon, a general practitioner in the U.K. and an honorary research fellow at Cardiff University in North Wales. He has been giving high-dose B vitamins to patients with elevated homocysteine levels for more than a decade, and has published case studies showing treatment benefits (see McCaddon et al., 2005 and McCaddon, 2006). McCaddon emphasized the importance of knowing your homocysteine level before considering B vitamin supplements, since supplements have no benefit at levels below 9.5 μmol/L.

However, Francine Grodstein at Brigham and Women’s Hospital, Boston, suggested caution when interpreting these results, pointing out that the VITACOG trial population was small, and therefore has less statistical power than large clinical trials. The overall body of evidence on the effects of B vitamin supplementation is inconsistent, Grodstein said, with many trials failing to find a benefit. In addition, she noted that in 1998, the U.S. instituted folate fortification of the food supply, yet in the years since, there has been no sign of decreases in the incidence of AD. “The existing evidence doesn’t support a tremendous benefit of folate supplementation,” she concluded, but added that it remains possible that certain groups of people may benefit more from B vitamin supplements than the general population does.

In the second paper, Gold and colleagues formally published the results of a six-month Phase 3 trial of rosiglitazone on 693 Alzheimer’s patients who were not taking any other insulin drugs or AD therapies. Rosiglitazone mediates insulin sensitization by acting on the peroxisome proliferator-activated receptor-γ (PPAR-γ). The drug had shown hints of benefit for AD patients in a pilot study and Phase 2 trial, although the effect seemed to depend on participants’ ApoE4 genotype (see ARF related news story on Watson et al., 2005; and ARF related news story on Risner et al., 2006). In the Phase 3 trial, however, the authors saw no significant difference in cognitive scores between those taking rosiglitazone and those taking placebo, regardless of genotype. Two other adjunctive studies also failed to find a treatment effect, Gold said (papers in press).

“GlaxoSmithKline’s investment in rosiglitazone for AD is pretty much finished,” Gold said, but notes that this class of drugs might still be worth pursuing by other companies. In PET scan studies, Gold said, GSK saw results that suggested that rosiglitazone does have an effect on cerebral metabolism, but the effect is modest and might take a long time to produce changes in the brain (paper in press). Six months might have been too short a trial to see results, Gold suggested. It is also possible, Gold said, that rosiglitazone did not enter the brain well enough to be effective, implying that related medications with better brain penetration might be more successful. Additionally, Gold pointed out that the Phase 3 trial included donepezil treatment as a positive control, but the authors saw a much weaker treatment effect from this approved AD drug than most other studies have shown, raising questions about the Phase 3 trial’s ability to show a treatment effect. Gold speculated that there might have been problems with GSK’s Phase 3 trial participant population, perhaps due to the difficulty of recruiting volunteers for monotherapy trials.

In related news, another PPAR-γ agonist, pioglitazone, successfully completed an 18-month safety trial in patients with probable AD. Of the 25 patients who completed the trial, 12 received pioglitazone and the rest got placebo. Reporting in the September 13 Archives of Neurology, lead investigator David Geldmacher, University of Virginia, Charlottesville, and colleagues write that the drug was well tolerated with few serious adverse events apart from peripheral edema, which is well known for this drug. Gold points out, however, that the sample size was too small to address concerns about relatively rare side effects. Although the safety trial results would allow drug development to proceed, “The negative rosiglitazone study steers us away from further pioglitazone studies in mild and moderate AD,” wrote Geldmacher in an e-mail to ARF. “Whatever effect the PPAR drugs have, it seems that they may be a day late and a dollar short for affecting progression in established cases of AD.” Geldmacher suggested that pioglitazone might be more effective in very early disease states such as MCI, but wrote that at this time they have no plans to move forward with the drug.—Madolyn Bowman Rogers


  1. This work in humans is nicely supported by studies performed in mice by a Japanese team earlier this year (incidentally, published in the same journal). The mouse work showed that vitamin B deficiency caused cognitive impairment, which could be blocked with an antibody against homocysteic acid, another cysteine metabolite (Hasegawa et al., 2010).


    . Treatment of Alzheimer's disease with anti-homocysteic acid antibody in 3xTg-AD male mice. PLoS One. 2010 Jan 20;5(1):e8593. PubMed.

    View all comments by Jürgen Götz

Make a Comment

To make a comment you must login or register.


News Citations

  1. Vienna: New Genes, Anyone? ICAD Saves Best for Last
  2. The B Side—Vitamins Won’t Sharpen AD Brains, Trial Suggests
  3. Orlando: Early Results Hint That Insulin-Sensitizing Drug Improves Cognition
  4. Madrid: Highs and Lows of The Insulin Connection

Paper Citations

  1. . Plasma homocysteine as a risk factor for dementia and Alzheimer's disease. N Engl J Med. 2002 Feb 14;346(7):476-83. PubMed.
  2. . A controlled trial of homocysteine lowering and cognitive performance. N Engl J Med. 2006 Jun 29;354(26):2764-72. PubMed.
  3. . Effect of 3-year folic acid supplementation on cognitive function in older adults in the FACIT trial: a randomised, double blind, controlled trial. Lancet. 2007 Jan 20;369(9557):208-16. PubMed.
  4. . A trial of B vitamins and cognitive function among women at high risk of cardiovascular disease. Am J Clin Nutr. 2008 Dec;88(6):1602-10. PubMed.
  5. . High-dose B vitamin supplementation and cognitive decline in Alzheimer disease: a randomized controlled trial. JAMA. 2008 Oct 15;300(15):1774-83. PubMed.
  6. . Co-administration of N-acetylcysteine, vitamin B12 and folate in cognitively impaired hyperhomocysteinaemic patients. Int J Geriatr Psychiatry. 2005 Oct;20(10):998-1000. PubMed.
  7. . Homocysteine and cognitive impairment; a case series in a General Practice setting. Nutr J. 2006;5:6. PubMed.
  8. . Preserved cognition in patients with early Alzheimer disease and amnestic mild cognitive impairment during treatment with rosiglitazone: a preliminary study. Am J Geriatr Psychiatry. 2005 Nov;13(11):950-8. PubMed.
  9. . 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.

External Citations

  1. New York Times story
  2. 18-month safety trial

Further Reading

Primary Papers

  1. . Homocysteine-lowering by B vitamins slows the rate of accelerated brain atrophy in mild cognitive impairment: a randomized controlled trial. PLoS One. 2010;5(9):e12244. PubMed.
  2. . Rosiglitazone monotherapy in mild-to-moderate Alzheimer's disease: results from a randomized, double-blind, placebo-controlled phase III study. Dement Geriatr Cogn Disord. 2010;30(2):131-46. PubMed.
  3. . A randomized pilot clinical trial of the safety of pioglitazone in treatment of patients with Alzheimer disease. Arch Neurol. 2011 Jan;68(1):45-50. PubMed.