Race and Genomics in Cardiovascular Research

1
Race and Genomics in Cardiovascular
Research Richard S. Cooper, MD Department of
Preventive Medicine, Loyola Stritch School of
Medicine Chicago, Illinois
2
Key Question Does continental race summarize
genetic variation of relevance to public health
or medicine?

3
Not, do races exist? Members of a subspecies
(ie, race) would share a set of phylogenetically
concordant phenotypic characters. Evidence
must normally come from the distributions of
multiple, independent, genetically based
traits. John Goodrum, 2003

4
Framework for Use of Race Categories How we infer
the role of genes in determining group variation.
. . Types of physical traits Group specific
A (eg, skin color, hair)

Correlated ? Variable B
(eg, hypertension, obesity) Invariant
C (eg, collagen, ATPase)

5
Health Status Measures in US Racial/Ethnic
Groups, 1998
White Black Hispanic Asian
Cause of Death Age
Adjusted Death Rates
450.4 121.9 79.2 23.3 121.0 21.9 12.7 12.0 2.6
690.9 183.3 92.5 41.4 161.2 17.7 17.4 28.8 20.6
432.8 84.2 54.7 19.0 76.1 8.5 9.8 18.4 6.2
264.6 67.4 42.9 22.7 74.8 7.4 10.3 8.7 0.8
All Causes Heart Disease Coronary Heart
Disease Stroke Cancer COPD Pneumonia /
Influenza Diabetes Mellitus HIV Infection
Infant Mortality (/1000) Life Expectancy
6.0 77.3
13.6 71.3
5.8 gt80?
5.5 gt80?
6
Cancer Mortality Rates by U.S. Racial/Ethnic
Groups, 1998
100
90
80
70
White
60
Black
Amerindian
50
Asian
40
Hispanic
30
20
10
0
Prostate1
Breast2
Colorectal1
Lung1
1men, 2women
7
The Race Factor LA Times, Sept 8, 2003
Researchers long believed that social,
environmental and economic stresses of
lower-income status explained (health
differentials) . . But they could not ignore a
growing mountain of research. Some racial
differences are encoded in the genes, and those
differences can make people of one skin color
inherently more or less susceptible to certain
diseases than people whose complexion is
different. In short, in matters of health,
race matters.
8
Mistake Number 1 Using the subtraction method
to quantify genetic effects.

9
Hypertension in HDFP, by Race and Education
50 40 30 20 10
Whites
Blacks
Hypertensive (DBP ? 95 mmHg)
0
lt 10 10-11 12 Some College years
years years College Grad
Educational Level
10
SES, Race and Mortality
Age-Adjusted Death Rate
SES vs. Death Rate
Blacks
Whites
Black

White
Education
11
Rates of Low Birthweight in Blacks Whites,
by Education Georgia , 1980 - 87
180 160 140 120 100 80 60 40 20 0
Blacks
Whites
Low Birth WT / 1,000
12
lt12
gt12
Years of Education
12
Rates of Low Birthweight in Blacks, by
Education Marital Status
200 180 160 140 120 100 80 60 40 20 0
Unmarried
Married
Low Birth WT / 1,000
12
lt12
gt12
Years of Education
13
Mistake Number 2 Assuming that population
prevalence can be interpreted as evidence of a
genetic effect.

14
Prevalence of Hypertension Among Six Populations
of West African Origin ICSHIB, 1995
0
.
4
0
0
.
3
5
0
.
3
0
0
.
2
5
Percent Hypertensive
0
.
2
0
0
.
1
5
0
.
1
0
0
.
0
5
0
.
0
0
M
a
y
w
o
o
d
J
a
m
a
i
c
a
S
t
.

L
u
c
i
a
B
a
r
b
a
d
o
s
N
i
g
e
r
i
a
C
a
m
e
r
o
o
n
15
Hypertension Prevalence In Europe and
North America
60
50.7
47.4
50
46.8
42.1
41.1
40
30
Prevalence of Hypertension in
24.3
21.8
20
10
0
USA
Canada
England
Finland
Germany
Spain
Italy
16
Age- and Gender- Adjusted Hypertension
Prevalence, by Country and Race
60
50
40
Hypertension Prevalence ()
30
20
10
0
Jamaica
Nigeria
US White
Sweden
Italy
England
US Black
Spain
Finland
Germany
Ages 35-64
17
Prevalence of Diabetes, by Race, Ages 20, 2000,
US
Prevalence ()
White 7.8 Black
13.0 Hispanic
10.2 Amerindian 15.1 (NIDDK)
18
Prevalence of Diabetes in Related Populations

Traditional Prevalence () Westernized
Prevalence() Amerindians Mapuche 0
Pima 23 New Guinea Rural 0
Urban 37 Australian Aborigines Rural 0
Urban 23 Middle East Yemen 4
Lebanon 14 Chinese Rural China 0
Urban Taiwan 13 Asian Indian Rural
India 0 Fiji 22 King and Rewers,
Diabetes Care, 1993
19
Prevalence of Diabetes by BMI and Sex in
Populations of West African Origin
22
20
18
16
14
12
UK
UK
Men
US
10
Women
Prevalence of Diabetes
US
Caribbean
8
6
Caribbean
4
West Africa
West Africa
2
0
20
22
24
26
28
30
32
Body mass index
20
Diabetes vs. Obesity in European-Origin
Populations
15
Women
Men
12
Ge
US
9
It
US
Ge
r 0.62
Diabetic
Sw
Ca
It
6
Ca
Sw
Ne
UK
De
Be
No
Be
No
3
UK
De
Ne
Fr
Fr
0
22
24
26
28
30
Mean BMI
Data WHO SURF, 2003, Self reported type 2
diabetes, persons 35-54
21

• Uses of Race in Public Health and Descriptive
  Epidemiology
• Surveillance
• Stratification 
• NOT etiologic analyses
•  

22
"A basic principle of epidemiologic studies is
that 'racial' or 'geographic' differences are not
adequate explanations for disease processes.
It is necessary to define and analyze those
environmental factors of habit, custom, marriage,
child-bearing, etc., which are peculiar to
population groups in these geographic areas."
Emerson Day, MD, speaking at the Second Biennial
Louisiana Cancer Conference, January, 1958  
23
(No Transcript)
24
Categorization of humans in biomedical research
genes, race and disease.


Risch N, et al. Genome Biology 20023/7/comment
2007.1-2007.12

• - Genetic studies have recapitulated the
  classical definition of races based on
  continental ancestry
• - Genetic differentiation is greatest when
  defined on a continental basis
• If biological is defined by susceptibility to
  and natural history of a chronic disease, than
  numerous studies over past decades have
  documented biological differences among races

25
Estimated World Population Structure Based on 52
Populations (Rosenberg,
Pritchard et al, Science, 2002)
26
Molecular Variance within and among World
Populations and Regions
Variance Components ()
No. of Regions 5 7
Within Pops 94.3 95.0
Among Pops, within Regions 2.3 2.3
Among Regions 3.4 2.8
Based on 274 tetranucleotide loci 52
populations Rosenberg et al, Science, June 2003
27
Regression Analysis of Individual Admixture on
Systolic BP among African Americans The NHLBI
Family Blood Pressure Program Predictive
Factors (Mean /- SE, N 981) Sys BP
Admixture Age Sex
BMI 114 -2.8 /- 3.7 0.2 /- .01
6.5 /- 0.8 0.4 /- .01   P lt
.001 Tang, Risch, Cooper, et al, submitted
28
Estimated Population Structure for Regions
29
Genetic Cluster Analysis, Regional Variation in
Common Disease and the Ecological Fallacy
The attempt to map disease rates onto
continental races is confounded by history and
the resulting structure within and between
regional world populations.

30
Using molecular genetics to study blackwhite
differences in common disease What is the
hypothesis?
Small et al. Synergistic polymorphisms of beta1-
and alpha2C-adrenergic receptors and the risk of
congestive heart failure. NEJM 20023471135
31
Alpha2 and Beta1 Adregergic Receptor
Variants and Heart Failure
Alpha2 Blacks- Controls Cases Whites
- Controls Cases Beta1 Blacks
- Controls Cases Whites- Controls Cases
Allele Frequency .41 .62 .04 .10 .56 .53 .76 .74
Odds Ratio 5.7 3.9 0.9 0.8
Small et al, NEJM, 2002
32
Gene-Gene Interactions of Alpha2 and Beta1
Adrengergic Receptor Variants and Heart Failure
N with genotype 15 2 2 3
N for Total Sample 78 84 81 105
Odds Ratio 10.0 2.1
Blacks - Cases Controls Whites
- Cases Controls
Small et al, NEJM, 2002
33
Gene Variants Tied to Heart Failure
in Blacks (Oct 10, Reuters Health)
Researchers have linked two common gene
variations to an increased risk of congestive
heart failure among black Americans. This
level of synergy has never been seen before in a
cardiovascular gene, Liggett noted. It is
possible, then, that these gene variations
contribute to black Americans particularly high
risk of CHF, he said.
34
Blacks, heart disease target of researchers
African-Americans may have genetic double-whammy.
AP, Oct 10, 2002.
A study published Thursday may help explain why
blacks are more likely to experience congestive
heart failure than whites.. Two pairs of genes
that rarely occur together in whites generate a
tenfold risk among black people who have both ..
Discrimination, lack of access to care and other
societal inequities are indeed a problem but may
not be the main reason for the gap, said Dr.
Yancy. When you control for those variables,
there is still more heart disease in
African-Americans. Particularly heart failure,
he said.
35
Genes and Variation in Disease Prevalence 1.
Single Gene Disorders Founder
effects restricted to sub-populations
Tay Sachs persons of Jewish descent
Cystic fibrosis northern Europeans
Selection defined by geography
Thalessemia extends from Italy to
Thailand Sickle cell much of
sub-Saharan Africa (but not South
Africa), the Middle East and India 2. Common
Complex Disorders Common to all populations
gene variants are shared world-wide Eg,
Hypertension lifetime incidence risk gt 80 in
the US
36
Hypotheses for Genetic Variation as a Cause of
Racial Variation in Common Disease
Common Common variants(? gt 10) are usually
old, unless under strong selection, and shared in
all groups. Many Rare Individually rare
alleles have a small effect on population risk.
The interactions required among rare alleles to
yield a large increase in population risk is also
implausible.
37
Pooled Odds Ratios for
Genetic Association Studies
Associated Gene, Number of Fixed Effects
Phenotypes Studies Odds Ratio ABCC8, exon
22, type 2 diabetes 4 2.28
COL1A1, osteoporotic fracture 12
1.59 CTLA4, type 1 diabetes 20
1.27 DRD3, schizophrenia 48
1.12 GSTM1, head/neck cancer 25
1.20 HTR2A, schizophrenia 28
1.07 PPARG, type 2 diabetes 14
1.22 SLC2A1, type 2 diabetes 3
1.76 Lohmueller et al, Nature Genetics,
Feb 2003
38
(No Transcript)
39
(No Transcript)
40
(No Transcript)
41
(No Transcript)
42
(No Transcript)
43
Race as a guide to drug therapy . . .
Genetic variants that alter drug metabolism vary
among populations . . . But virtually never in an
all-or-none pattern. For race to be a surrogate
for a variant, you need high (gt 90) sensitivity
AND specificity. If you want to know whether a
patient has a variant you have to test for it.
44
Race as a guide to drug therapy . . .
Variation in Cytochrome P450 Gene Variants
Populations Allele Frequency ()
Gene A B C D CYP1A2 34 31
40 41 CYP2C19 9 37 27
25 CYP2D6 53 39 70 30 Wilson
et al, Nature Genetics 2001
45
Whose Genome Is It After All? Studying
Differences vs. Studying Similarity
46
The Jeffersonian
Genome
Observed Trait
Genome
? IQ ? BP
47
The Anthropologists Genome
Cavalli-Sforza, The Great Human Diasporas
48
(No Transcript)