Lever 1
Physical exercise, the most thoroughly documented lever
Among all non-pharmacological interventions studied against cognitive decline, physical exercise remains the most solidly validated by research.
A UK Biobank study of nearly 90,000 adults, followed on average for 4.4 years, precisely measured the relationship between the dose of moderate-to-vigorous activity and the risk of dementia: −41 % from 35 minutes per week, −60 % between 35 and 70 minutes, −63 % between 70 and 140 minutes, and −69 % above 140 minutes per week, compared to no activity. Each extra 30-minute slot per week is associated with around 4 % less risk.
The most encouraging message: there is no discouraging minimum threshold. Even very small amounts of activity are associated with strong risk reductions, which makes this strategy accessible even to the most sedentary or fragile.
The mechanism: how exercise protects the brain
The protective effect is not magical. It runs through a molecular cascade that is increasingly well understood.
BDNF. BDNF (brain-derived neurotrophic factor) is the central molecule. It drives neurogenesis in the hippocampus, the key memory region, and supports synaptic plasticity and neuron survival. Its decline is considered an early marker in Alzheimer's disease.
Irisine. Irisin, a myokine released by muscles during effort, crosses the blood-brain barrier and stimulates BDNF production in the hippocampus. This is one of the mechanisms that explains how muscle activity 'speaks' directly to the brain.
Neurogenesis. Adult hippocampal neurogenesis (AHN) completes the picture. A 2018 Science study showed, in Alzheimer's mouse models, that neither stimulating neurogenesis alone nor exercise alone (without an increase in this neurogenesis) improved cognition. Both mechanisms must be activated together. Animal-model work, but it laid the conceptual foundation for several current human research tracks.
Lever 2
Sugar, glycation and 'type 3 diabetes'
Chronic hyperglycaemia is not only a metabolic problem: it has a direct impact on the brain. When excess sugar circulates in the blood over time, it favours the formation of advanced glycation end products (AGEs), which accumulate in tissues, including brain tissue.
These AGEs interact with a specific receptor, RAGE, and with the TLR4 receptor, triggering a cascade that includes oxidative stress, mitochondrial dysfunction, neuroinflammation, plaque deposition and neuronal death. This mechanism helps explain why Alzheimer's disease is sometimes nicknamed 'type 3 diabetes': the brain of a person with chronic hyperglycaemia develops a form of localised insulin resistance that worsens the pathological processes.
At the molecular level, RAGE activation triggers pathways involving GSK-3β and MAP kinases, which promote abnormal phosphorylation of the Tau protein, one of the two major histological hallmarks of Alzheimer's alongside amyloid plaques.
Practical implication: limiting dietary glycation sources (high-temperature cooking, grilling, frying, ultra-processed foods) and stabilising daily blood sugar is not only diabetes prevention. It is a neuroprotection strategy in its own right.
Lever 3
Genetic terrain: ApoE and MTHFR
Lifestyle acts differently depending on each person's genetic terrain. Two polymorphisms are particularly well documented in this context.
ApoE e4
The ApoE gene exists in several forms (alleles); the e4 variant is the most solidly established genetic risk factor for Alzheimer's disease, present in about 20 % of the population. Heterozygous carriers (one copy) have a risk multiplied by 3 to 4; homozygous carriers (two copies) by 9 to 15, with an onset earlier by 5 to 10 years on average.
This genotype is not a fatality: it is precisely the terrain on which lifestyle seems to have the most leverage, which makes it an even higher priority for carriers, not a reason for fatalism.
MTHFR C677T
This polymorphism affects a key enzyme of the methylation cycle. In homozygous carriers of the 677T variant, enzyme activity is reduced by around 70 % compared to the wild type (heterozygotes sit around 65 % residual activity). This translates into elevated blood homocysteine, an amino acid whose excess is associated with damage to the cerebral vascular endothelium and insufficient methylation, notably of myelin and some neurotransmitters.
This variant is nutritionally correctable, through suitable intake of folate (in its methylated form for carriers of the variant), B12 and B6, which normalise homocysteine levels.
Bottom line
The brain is not a frozen organ, and cognitive risk is not a fate written in genes alone. Three main levers act in synergy:
- Regular physical exercise, working through the irisin-BDNF-neurogenesis axis, with measurable benefit from small weekly doses.
- Blood-sugar and dietary glycation management, to limit the AGE-RAGE-neuroinflammation cascade.
- Knowing your genetic terrain (ApoE e4, MTHFR), so you can adjust prevention priorities rather than face an unidentified risk.
Other polymorphisms such as COMT, DIO2 and FADS1/FADS2 also play a role in cognitive health.
⚕️ In France, only a medical doctor is authorised to prescribe a genetic test.
👉 A naturopath can help adapt these principles to your genetic terrain and lifestyle.
Sample plates
For ApoE3/4 and ApoE4/4 carriers
For people carrying at least one copy of the ApoE4 allele, certain dietary adjustments are particularly well documented. The recommendations below are based on what research has specifically studied in this genetic profile, not on generic 'brain-healthy' nutrition advice.
What studies on ApoE4 carriers show
A Mediterranean-style pattern remains the best-documented baseline, with data suggesting a particularly protective effect in ApoE4/4 homozygotes showing signs of metabolic dysfunction (insulin resistance, unfavourable lipid profile). ApoE4 carriers have a globally less favourable lipid profile and seem more sensitive to saturated fat and dietary cholesterol: limiting them is therefore a sharper priority than in the general population. Because lipid transport is altered by ApoE4, a higher and more regular intake of omega-3s (DHA in particular) may be necessary to reach sufficient brain levels.
Important: the ketogenic diet is not recommended for this profile. Contrary to a sometimes widespread idea, current data suggest that ApoE4 carriers benefit less from, or not at all from, the ketogenic diet, probably due to a mitochondrial function already impaired by this variant. It is therefore not a priority option here.
Sample plate: lunch
- A serving of oily fish (sardines, mackerel, salmon) in olive oil or steamed, to preserve omega-3s as well as possible.
- Leafy greens (spinach, rocket, chard) and varied colourful vegetables.
- A serving of pulses (lentils, chickpeas) for fibre and moderate glycaemic load.
- Dressing with extra-virgin olive oil, rich in polyphenols.
- A small handful of nuts or almonds.
Sample plate: dinner
- A lean protein (poultry, eggs) or a second portion of oily fish during the week.
- Wholegrains in moderate amounts (quinoa, brown rice) rather than refined ones.
- Cruciferous vegetables (broccoli, cauliflower) on the side.
- A few berries (blueberries, raspberries) for dessert, rich in antioxidants.
- Keep butter, processed meats, very fatty cheeses and ultra-processed products to a minimum.
A nuance: the Mediterranean diet's protective effect on cognitive risk varies across populations (more marked in some North American and Mediterranean cohorts than in Nordic ones, for example), and part of these recommendations rests on understanding ApoE4 mechanisms rather than on randomised clinical trials run specifically in human carriers of this variant. Many other factors also matter: physical activity, sleep, other biomarkers, individual digestive tolerance. These examples are a starting point, not a prescription.
Go further
