The idea that antioxidants might help people with AD finds support from the scientific literature. Epidemiology studies suggest that diets rich in antioxidants reduce AD risk (Morris, 2009). In AD mouse models, supplementation with vitamin E, a powerful antioxidant, reduces Aβ pathology, preserves neurite health, and improves cognition (see Sung et al., 2004; Garcia-Alloza et al., 2006; Harrison et al., 2009). Before launching the current study, the Alzheimer’s Disease Cooperative Study (ADCS) conducted two antioxidant trials—one in moderate to severe AD patients (Sano et al., 1997), the other in people with mild cognitive impairment (ARF related news story on Petersen et al., 2005). Neither found cognitive benefit with treatment, though some of the AD patients showed functional improvement. Scientists puzzled over how to interpret the data—were the doses high enough, would certain combinations of vitamins work better, etc.? At the same time, a meta-analysis of 19 randomized trials warned that high doses of vitamin E, that is, greater than 400 international units (IU) per day, may be linked to increased mortality (see Miller et al., 2005). But after all these studies, a critical question still remained. “We weighed all these things, took a step back, and decided that we actually didn’t know how well vitamin E (or other antioxidants) reduced oxidative stress in the brain,” Galasko told Alzforum. “So we decided to do a biomarker study to investigate.”

Galasko and colleagues enrolled 78 AD patients at 12 ADCS sites. They looked for change in potential CSF biomarkers of AD (Aβ42, tau, and phospho-tau₁₈₁) and oxidative stress (F2-isoprostane)—as well as cognition (Mini-Mental State Exam) and daily function (ADCS Activities of Daily Living Scale)—at the end of the 16-week treatment period. One-third of the participants were randomized into a placebo group. The remainder received one of two treatments—a daily supplement of 800 IU vitamin E, 500 mg vitamin C, and 900 mg α-lipoic acid (ALA); or a chewable wafer containing 400 mg coenzyme Q (CoQ), a naturally occurring antioxidant that protects mitochondria from oxidative stress. A similar combination of vitamins E and C and ALA (E/C/ALA) improved skeletal muscle metabolism in healthy seniors (Wray et al., 2009), and the CoQ dose had improved function in some patients in a large Phase 2 trial of Parkinson’s disease (Shults et al., 2002).

None of the AD biomarkers budged with either treatment. CSF levels of F2-isoprostane fell about 19 percent in the E/C/ALA group, suggesting that this antioxidant combination did what it was supposed to, that is, lower oxidative stress in the brain, albeit modestly. “This is a vast improvement over other antioxidant trials where no one knows if the antioxidants are doing anything,” noted Greg Cole of the University of California, Los Angeles. Nevertheless, Galasko said, it is hard to know if the 19 percent reduction in F2-isoprostanes is biologically meaningful. Whereas studies suggest that a 25-30 percent drop in Aβ levels can slow pathology in AD transgenic mice, “we didn’t have a similar framework for evaluating antioxidants,” Galasko noted.

In contrast, CoQ did not even change CSF F2-isoprostane levels. Again, it is difficult to know what to make of that, Galasko said. The study did not measure how well the antioxidants entered the brain. And for CoQ, “it’s not only a question of getting in, but reaching mitochondria where it probably needs to be. It’s not clear how to best measure that,” Galasko said. Furthermore, many biomarkers are global readouts of changes occurring in small compartments, which can further complicate analysis, noted George Perry of the University of Texas at San Antonio.

However, despite the challenges of interpreting biomarker data, “looking at it at face value, we do not suspect that the antioxidants altered the AD process during the four-month period in this study,” Galasko said.

As for the cognitive data, the E/C/ALA group seemed to decline faster. This was suggested by a 2.8 point change on MMSE scores from baseline, compared to 0.9-1.0 point changes in the placebo and CoQ groups. “We did not power the study to detect a valid cognitive signal in four months,” Galasko said. Hence, it is hard to say whether the MMSE decline “was a chance occurrence or if it was some effect of the drug combination,” he told Alzforum. The E/C/ALA formulation has been used in clinical trials for diabetes and other conditions, but those did not include cognitive monitoring, Galasko noted. Nevertheless, the apparent cognitive setback raises a flag, he said. “It’s something that would need to be looked at carefully in further studies.”

Mary Sano of Mount Sinai School of Medicine, New York, found the study convincing. “I completely believe these data,” she said in an interview with Alzforum. “It says that antioxidants do not have a benefit on cognition, and may have a detrimental effect. This is paralleled by a decrease in isoprostanes. It’s a very powerful study.”

The findings challenge the simplistic belief that loading up on vitamins and antioxidants is good for you. “We like to say antioxidants are great and do wonderful things,” Sano said. “But the data are weak and may be contradictory. We have to come to grips with that.” Though the 1997 ADCS study showed that antioxidants improved function in AD patients, the supplements failed to influence cognition in several large prevention trials (Yaffe et al., 2004; Heart Protection Study Collaborative Group, 2002), making it hard to judge whether antioxidants do much good for brain health.

“It is certainly possible for an apparently innocuous cocktail of ingredients to work poorly together,” Cole noted. “The real lesson is that designing antioxidant cocktails is something that shouldn't be left to the supplement makers. They need to be rigorously tested.”

In follow-up, Galasko and colleagues are looking at biological samples from ongoing longitudinal studies to discover plausible correlations between plasma and CSF measures, and reported use of supplements. Galasko said his team does not have plans to test antioxidants in milder AD or at-risk, asymptomatic populations.—Esther Landhuis.

Reference:
Galasko DR, Peskind E, Clark CM, Quinn JF, Ringman JM, Jicha GA, Cotman C, Cottrell B, Montine TJ, Thomas RG, Aisen P for the Alzheimer’s Disease Cooperative Study. Antioxidants for Alzheimer Disease: A Randomized Clinical Trial with Cerebrospinal Fluid Biomarker Measures. Arch Neurol. 19 Mar 2012. Abstract