APPGD 27 November 2013

Modifying risk factors for dementia: where are we now and where do we need to go?

A. David Smith, University of Oxford

What progress has been made in identifying potential risk factors for dementia (mainly Alzheimer’s disease)?

• Several genetic risk factors, not modifiable
• Many non-genetic risk factors postulated, few proven. Best evidence for:
─ Low education
─ Smoking
─ Mid-life high blood pressure, high cholesterol and obesity
─ Depression
─ Physical inactivity
─ Diabetes & high blood sugar
─ Low intake of omega-3 fatty acids
─ High blood homocysteine (due to low-normal B vitamin status)
• Modifying non-genetic risk factors can prevent disease, e.g. heart disease has declined by 70%, half of decline due to vascular risk factor modification
• Since vascular risk factors are also linked to dementia, has prevalence of dementia also declined: yes, in the UK it appears to have declined by 24% in last 20 years (MRC CFAS study)
• About half of all Alzheimer’s disease cases world-wide might be attributable to known nongenetic risk factors

An update on clinical trials that test modifying risk factors

• Most trials have been done on people with dementia and have failed; need trials at an earlier stage, such as mild cognitive impairment (MCI)
• Pilot trials in USA and Germany on omega-3 fatty acids have shown benefit
• Pilot trials in Australia and USA on exercise have shown benefit
Pilot trial on homocysteine-lowering by B vitamins in UK has shown a marked slowing of disease progression (see description further below)

Where do we need to go now? Are there any public health implications?

• Urgent need for more funding of dementia research overall, in particular prevention
• UK has poor record of supporting dementia prevention research. Only 0.1% of Research Council expenditure on dementia since 2006 was for prevention; none from NIHR since 2000
• Ways of promoting the funding of prevention research should be considered
• We need a cultural change in attitudes: instead of almost exclusively focussing dementia research on unravelling disease mechanisms and genetics, we need a more holistic attitude in which the search for risk factors plays an important role
• Identify subgroups of the population at risk, using a simple screening test
• Trials to modify risk factors should be done in two stages:
─ A pilot trial in one centre on people with MCI in which the main end-point is the slowing of brain atrophy; this may help to identify a sub-group that will benefit
─ A larger multicentre trial in MCI, ideally in the vulnerable sub-group, in which the main endpoint is conversion to dementia
• Adopt public health policies based upon the trial results:
─ screening for those at risk, using biomarkers of the risk factors, (e.g. homocysteine)
─ followed by appropriate treatment (e.g. B vitamins)
─ possible public health intervention (e.g. fortification of food with a vitamin)
─ develop ways of encouraging changes in population behaviour towards a better life-style

Evidence suggests that we may be able to reduce the incidence of dementia considerably in the next 15 years, so saving much suffering and a large sum of money. 

Example: a trial of high-dose B vitamins in Mild Cognitive Impairment (VITACOG) Douaud, G., et al. Proc Natl Acad Sci U S A (2013) 110: 9523-9528.

• Raised blood levels of the amino acid homocysteine are associated with an increased risk of developing cognitive impairment and later, Alzheimer’s disease
• The main determinant of blood homocysteine is the body’s status of 3 of the B vitamins (folate, vitamins  B6 and B12). Homocysteine levels can be lowered by administering these vitamins
• 270 people over 70 with Mild Cognitive Impairment living in the community were recruited. MRI scans were done at recruitment and half participants were given high doses of the 3 B vitamins, half took a placebo tablet. After 2 years, everyone was scanned again
• The two MRI scans were analysed to find the rate of shrinkage (atrophy) of the whole brain
• The rate of brain atrophy in those treated with the B vitamins was on average 30% slower than those taking placebo
• The effect of B vitamin treatment was more marked in those whose blood homocysteine level was highest at baseline; in these people the B vitamins slowed the atrophy by 53%
• At the same time, the rate of decline in several cognitive tests, including memory, was markedly slowed by the B vitamins in those with high starting homocysteine
• The brain scans were analysed in more detail to see which parts of the brain were protected by the B vitamin treatment. The result is shown below; yellow indicates areas with atrophy after 2 years




• In the B vitamin group there was a 90% slowing of brain atrophy in the regions highlighted
• These are the same regions that are atrophied in Alzheimer’s disease
Thus, it is possible markedly to slow down the disease process in people with raised blood homocysteine levels
• About half the elderly population have such raised levels of homocysteine
• Screening for blood homocysteine would identify those at risk of increased brain atrophy and eventual dementia
• They could be offered treatment with high-dose B vitamins, which are considered safe

This trial shows the feasibility of prevention trials, the importance of selecting a vulnerable subgroup, and establishes the principle that the Alzheimer’s disease process can be slowed down by modifying a risk factor.


Some key references to the literature:

The projected effect of risk factor reduction on Alzheimer's disease prevalence. Barnes, D. E. and K. Yaffe. Lancet Neurol (2011) 10: 819-828.

Physical exercise as a preventive or disease-modifying treatment of dementia and brain aging. Ahlskog, J. E., et al.. Mayo Clin Proc (2011) 86: 876-884.

Mediterranean diet, cognitive function, and dementia: a systematic review. Lourida, I., et al. Epidemiology (2013) 24: 479-489.

Long-chain omega-3 fatty acids improve brain function and structure in older adults. Witte, A. V., et al.. Cereb Cortex (2013) June, Epub ahead of print.

Preventing Alzheimer's disease-related gray matter atrophy by B-vitamin treatment. Douaud, G., et al. Proc Natl Acad Sci U S A (2013) 110: 9523-9528.

Nonpharmacologic treatment and prevention strategies for dementia. Yaffe, K. and T. Hoang. Continuum (Minneap Minn) (2013) 19: 372-381.

A two-decade comparison of prevalence of dementia in individuals aged 65 years and older from three geographical areas of England: results of the Cognitive Function and Ageing Study I and II. Matthews, F. E., et al., on behalf of the Medical Research Council Cognitive Function and Ageing Study. Lancet (2013). 10.1016/S0140-6736(13)61570-6. Epub ahead of print.

The worldwide economic impact of dementia 2010. Wimo, A., et al. Alzheimer Disease. Alzheimers Dement (2013) 9: 1-11 e13.