Pascale Barberger-Gateau, PhD, Centre de recherche INSERM, Université de Bordeaux, France
“In accordance with the results of basic research, epidemiological studies suggest a protective effect of several classes of nutrients against cognitive decline and risk of dementia (1, 2), including principally long-chain omega-3 polyunsaturated fatty acids (n-3 PUFA) (3, 4), vitamins C and E (5), carotenoids (6) and B vitamins (7), although some discordant data exist as well. Many underlying mechanisms can be evoked to support biological plausibility of a protective effect of these nutrients against brain aging: role in brain composition and function, hippocampal neurogenesis, limitation of accumulation of the beta-amyloid peptide, decreased oxidative stress and inflammation, decreased homocysteine concentration and vascular effects. However, most randomized controlled trials (RCTs) with nutritional supplements have yielded disappointing effects on cognition so far.
RCTs with nutritional supplements have been carried for the primary or secondary prevention of dementia or for treatment of patients with mild-to-moderate Alzheimer’s disease (AD). Some may have failed to show any significant impact on cognition, either because they have included healthy participants who did not decline during the few months of the trial or, conversely, individuals with AD whose disease was probably too severe to expect a significant effect from nutrition. Few studies have specifically targeted the mild cognitive impairment (MCI) stage. Identifying individuals who are the most susceptible to benefit from a nutritional supplementation is therefore of upmost importance (8). In its strict definition, primary prevention takes place before thedevelopment of the disease, that is, before any sign of neurodegeneration in the case of AD. Healthy diets in early life could contribute to develop brain reserve. However, the impact of interventions targeting this population on late life cognitive performance is obviously impossible to evidence. Thus primary and secondary prevention of dementia cannot be disentangled. However, cognitive decline is a relatively late event in the course of the disease, and it has been estimated that beta-amyloid, the hallmark of AD, accumulates sharply for about 15 years in the brain before reaching a plateau when the first clinical signs emerge. This 15-year interval would represent a large window for secondary prevention, especially with nutritional interventions. In the absence of specific impairment on most neuropsychological tests at this stage, biomarkers of disease progression could help to identify the best window of opportunity for RCTs with nutritional supplements. These biomarkers include brain atrophy as measuredby magnetic resonance imaging, impaired brain glucose metabolism and imaging of beta-amyloid accumulation with positron emission tomography.
Contrarily to RCTs with drugs, participants in nutritional interventions do not have a null basal level of the nutrient to be evaluated before any supplementation. There is a large day-to-day (intraindividual) variability, especially for nutrients brought by foods that are not consumed on a daily basis such as fish, the maindietary provider of eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA). In addition, there is a large interindividual variability, depending on dietary habits. Providing individuals who already meet their dietary requirements with additional quantities of nutrients is probably useless and may even be harmful when upper tolerable intake levels are surpassed. Inclusion criteria should consider nutritional criteria such as lowintake of EPA or DHA (9-11) or low blood levels of long-chain n-3 PUFA (4).
Gene-diet interactions on cognition are still poorly understood. The epsilon 4 allele of the Apolipoprotein E (APOE4) gene, the main genetic risk factor for AD, is associated with poorer blood response to supplementation with n-3 PUFA or fish intake. Moreover, some epidemiological studies and RCTs have evidenced that APOE4 carriers had no benefit of n-3 PUFA supplementation or fish consumption on cognition (12). Other genetic risk factors of AD involved in lipid metabolism could also interact with dietary n-3 PUFA (13). Further research is needed to better identify these interactions before considering genetic characteristics as inclusion criteria in RCTs with nutritional supplements.
Inability of most RCTs with nutritional supplements to show any effect on cognitive performance may in part lie in lack of sensitivity of the selected outcomes to change over a relatively short period. Beside traditional criteria based on cognitive and functional decline, biomarkers could contribute to provideevidence of an impact of nutrients and shed new light into their mechanisms of action. Few RCTs with nutritional supplements have biomarkers of disease progression (e.g., neurodegeneration) as objectives. These biomarkers could be more sensitive and capture early changes in the course of the disease that do not translate yet into decreasing cognitive performances. Many biomarkers of putative mechanisms of action of nutrients can also be proposed in RCTs. These include principally measures of inflammation, oxidative stress, homocysteine and markers of insulin resistance.
Despite the disappointing results of nutritional interventions on cognition so far, there is considerable room for improvement and more evidence-based knowledge on the link between nutrition and cognitive decline in older persons. Trials with nutritional supplements should learn from RCTs with drugs regarding biomarkers of disease progression for inclusion criteria and outcomes. Progress in genetics and omics technologies will also offer new opportunities for research in this field.
In addition, it may make sense to investigate nutrients, which previous studies have neglected and that could have an important role in the brain. As EPA and DHA, the omega-3 fatty acid docosapentaenoic acid (DPA) is mainly found in fatty fish and could contribute to the apparently protective effect of fish consumption against cognitive decline. The metabolism and biological functions of DPA are still poorly understood. Supplementation with pure omega-3 DPA induced a significant rise in the proportions of EPA and DHA in plasma triglycerides, suggesting that DPA may act as a reservoir of the major long-chain omega-3 PUFA in humans (14). Beyond its role in bone health, the multiple biological functions of vitamin D are raising increasing interest. In the Women’s Health Initiative study, supplementation with 400 IU/day of vitamin D, along with calcium, did not result in decreased risk of dementia or cognitive decline, compared with the placebo group over a mean follow-up of 8 years (15). If ongoing RCTs yield more positive results with higher doses of vitamin D, improvement of cognitive function would be a beneficial side-effect of systematic vitamin D supplementation in deficient individuals. A growing body of evidence supports a role for vitamin K in brain and cognition (16). Several vitamin K-dependent proteins contribute to brain function. Preliminary data need to be confirmed in longitudinal studies using serum phylloquinone as a biomarker of vitamin K status.
Moreover, most previous epidemiological studies and RCTs have focused on single nutrients, ignoring their potential synergistic effects when they are provided in optimal quantities and proportions as in a balanced diet. Some combinations of nutrients may have synergistic effects that will reinforce their individual properties. Considering healthy dietary patterns instead of single nutrients may lead to a more holistic approach of the diet (17). Also, supplements containing various combinations of nutrients trying to reproduce healthy diets can be used to provide a proof of their positive impact on cognition in RCTs (18).”
Based on: Barberger-Gateau P. Nutrition and brain aging: how can we move ahead? European Journal of Clinical Nutrition. 2014; 68:1245–1249.