dementia: genetics and lifestyle both matter

A large study found that genetic risk and lifestyle both played a role in the development of dementia, and that their roles were independent of each other (see dementia lifestyle dec risk w genetic risk jama2019 in dropbox, or JAMA. doi:10.1001/jama.2019.9879)

Details:

--retrospective study of adults of European ancestry without baseline cognitive impairement who were in the UK Biobank, a population-based cohort of >500,000 people, from 2006-2010

--196,383 people were followed for 8.0 years, with 1,545,433 person-years of followup

--mean age 64, 47% male, 42% had higher education, 60% in the middle 3 SES quintiles, 92% not current smokers, 77% regular physical activity, 51% healthy diet, 56% moderate alcohol consumption, 4% history of stroke, 18% depressed in the past 2 weeks

    --of these demographics, more in the group with dementia were: male (55 vs 47%), had higher education (42 vs 31%), were in the lowest SES (28 vs 20%), were depressed in the past 2 weeks (25 vs 18%)

--a favorable lifestyle was defined as (all scored 1 point, for a total of up to 4):

    --smoking status: current vs not

    --regular physical activity: as defined by the AHA: >150 minutes of moderate activity/week, >75 minutes of vigorous activity/wk, or engaging in moderate activitay 5 d/week or vigorous activity 1 d/wk

    --diet: consumption of at least 4 of 7 commonly eaten food groups, per recommendations of AHA: fruit 3 servings/d; nuts/seeds 4 servings/wk; veges including legumes 3 servings/d; whole grains 3 servings/d; fish/shellfish at least 2 servings/wk; dairy 2-3 servings/d; vegetable oils 2-6 servings/d (see https://www.ahajournals.org/doi/full/10.1161/CIRCULATIONAHA.115.018585 ). studies have shown that this diet is associated with better late-life cognition and reduced dementia risk

    --moderate alcohol consumption: 0-14g/d for women or 0-28 g/d for men.  some prior studies have suggested a U-shaped curve between alcohol use and dementia

--classification of lifestyle: favorable if 3-4 of these factors, intermediate if 2, unfavorable if 0-1

    --a weighted risk score was derived, including adjustment for age, sex, education, SES, 3rd-degree relatedness, and first 20 principal components of ancestry

--polygenic risk for dementia: multiple genetic risk factors, including apolipoprotein E4, mutations in amyloid prcursor protein, presenilin 1,and 2, and others

    --high genetic risk was those in the highest quintile, intermediate for quintiles 2-4, and low in those in the lowest quintile

Results:

--lifestyle: favorable in 68% of population, intermediate in 24%, and unfavorable in 8%

--1769 incident cases of all-cause dementia

--according to genetic profile:

    --high risk: 1.2% developed dementia

    --low risk: 0.6% developed dementia

        --adjusted Hazard Ratio, comparing high vs low genetic risk 1.91 (1.64-2.23)

--according to lifestyle:

    --unfavorable: 1.2% developed dementia

    --favorable: 0.8% developed dementia

        --adjusted Hazard Ratio, comparing high vs low genetic risk 1.35 (1.15-1.58)

--comparing the group with high genetic risk and unfavorable lifestyle to those with low genetic risk and favorable lifestyle: 1.78% (1.38-2.28%) developed dementia vs 0.56% (0.48-0.56%); HR 2.83 (2.09-3.83)

--in the group with high genetic risk: there was consistently was about 2x the dementia risk of those with low genetic risk; this was independent of the lifestyle component

--the lifestyle component increased 50% going from favorable to unfavorable, and it did so in both the high and low genetic risk groups (ie the same relative numbers applied to those in these different genetic risk groups, with the dementia risk increasing 50% going from favorable to unfavorable)

    --ie, there was no significant interaction between lifestyle and genetic risk

--no significant difference by several sensitivity analyses: using unweighted lifestyle scores, adjusting for depressive symptoms, with more alcohol consumption, or when results stratified by age, sex, and educational level

    --though there was a higher demetntia risk in those with a stroke history, with HR 2.74 (2.03-3.71) for those with high genetic risk and unfavorable lifestyle vs low genetic risk and favorable lifestyle

Commentary:

--this study adds to prior studies finding a healthy lifestyle is associated with decreased dementia risk (eg see http://gmodestmedblogs.blogspot.com/2015/12/lifestyle-interventions-and-cognition.html ; http://gmodestmedblogs.blogspot.com/2017/05/sugary-beverages-and-alzheimers.html ; http://gmodestmedblogs.blogspot.com/2015/11/mediterranean-diet-and-brain-volume.html )

--in the young cohort in this study, the absolute risk reduction for those in the high genetic risk group with favorable vs unfavorable lifestyle was 0.65%, translating to 1 case of dementia prevented for each 121 individuals per 10 years who improved their lifestyle, per the investigators (though this is a bit of a leap: this study was not an RCT with an intervention to change lifestyle and see the resultant dementia incidence!!)

--this study confirms that there is a gradient of dementia risk related to genetics as well as to lifestyle. And these are independent of each other. interestingly, the absolute risk for those with low genetic risk but unhealthy lifestyle (29/24460=0.12%) is remarkably close to those with a high genetic risk but a favorable lifestyle (298/208769=0.14%)

--other studies from the UK Biobank:

    --a study of lifestyle vs genes for strokes, also finding the each played an independent role: http://gmodestmedblogs.blogspot.com/2018/11/lifestyle-vs-genes-in-who-has-strokes.html

    --a study finding that coffee was associated with decreased mortality, independent of genetic variations in caffeine metabolism: http://gmodestmedblogs.blogspot.com/2019/01/coffee-and-decreased-mortality.html

    --of note, a study from a Chinese biobank found that there was no cardiovascular benefit from moderate alcohol intake (stroke risk was lowest if zero alcohol intake): http://gmodestmedblogs.blogspot.com/2019/05/stroke-risk-lowest-if-zero-alcohol.html . my guess is that the role of alcohol in dementia is actually that zero is better than moderate.... (perhaps through another Biobank study)

--there were several limitations to this study:

    --the lifestyle score has not been independently validated (and it is pretty likely that assigning a binary attribution raises a few issues: are there actually the same attributable risk of eating well as moderate alcohol consumption?? and does it make sense to consider smoking 1 cigarette/day as the same attributable risk from smoking 2 packs a day??, or that non-smoking includes those with 80 pack-years of smoking but stopped last week?)

    --there were only 8 years of followup in a pretty young cohort, with the average age of 72 at the end of the study (which might explain why only a couple of % developed dementia in the course of the study). longer-term outcomes might have been different

    --this study can only note associations and not causality: was there an unexpected cofounder present, not taken into account, which explains the association (eg other lifestyle/life conditions issues not included in the 4 above that was the most important one)??  also, are we really sure that the SNPs identified are the only ones associated with dementia and that an unidentified one now may really be more important. or might there be more complex associations (eg one SNP is associated with dementia perhaps only in the context of another one also found, or even one is associated only in the absence of another)

    --the group selected (white UKers) limits the generalizability of the results

--other studies have challenged the issue of genetic determinism:

    --see http://gmodestmedblogs.blogspot.com/2017/04/23andme-genetic-analysis-approved-for.html on FDA approval of direct-to-consumer 23andMe genetic analysis, which as a test has quite low sensitivity, specificity and positive predictive values, but conceptually reinforces to people that genetics are determinant.

     --http://gmodestmedblogs.blogspot.com/2018/01/dietary-effect-strong-when-high-genetic.html found that dietary changes were actually more effective in those with a high genetic predisposition to weight gain

so, another study finding that genetics are not determinant for another common condition, and that lifestyle really matters. i am concerned that the increasing studies/popular information/new therapies involving genetics are increasingly transforming the focus to genetics at the expense of lifestyle and people's conditions (poverty, income inequality, stress, diet/exercise and true access to healthy diets/exercise, exposure to carcinogenic materials/infections in the environment), and that in most of the prevalent chronic conditions, even ones with a pretty clear genetic component (eg obesity, diabetes, hypertension...) have a very strong lifestyle component that, if reinforced, can potentially negate the effect of genetics....

geoff

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@uphams.org

to get access to all of the blogs:

1. go to http://gmodestmedblogs.blogspot.com/ to see them in reverse chronological order

2. click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category

3. or you can just click on the magnifying glass on top right, then  type in a name in the search box and get all the blogs with that name in them