Download Genome-Wide Association Studies

ME62CH02-Hirschhorn
ARI
ANNUAL
REVIEWS
4 December 2010
Further
Annu. Rev. Med. 2011.62:11-24. Downloaded from www.annualreviews.org
by University of Oxford - Bodleian Library on 05/27/11. For personal use only.
Click here for quick links to
Annual Reviews content online,
including:
• Other articles in this volume
• Top cited articles
• Top downloaded articles
• Our comprehensive search
11:3
Genome-Wide Association
Studies: Results from the
First Few Years and Potential
Implications for
Clinical Medicine
Joel N. Hirschhorn and Zofia K. Z. Gajdos
Department of Genetics, Harvard Medical School, Boston, Massachusetts 02115; Program
in Genomics and Divisions of Genetics and Endocrinology, Children’s Hospital, Boston,
Massachusetts 02115; Broad Institute, Cambridge, Massachusetts 02142;
email: [email protected], zofi[email protected]
Annu. Rev. Med. 2011. 62:11–24
The Annual Review of Medicine is online at
med.annualreviews.org
This article’s doi:
10.1146/annurev.med.091708.162036
c 2011 by Annual Reviews.
Copyright All rights reserved
0066-4219/11/0218-0011$20.00
Keywords
human genetics, common disease, quantitative traits, personalized
medicine, biological pathways, common variants
Abstract
Most common diseases and quantitative traits are heritable: determined
in part by genetic variation within the population. The inheritance is
typically polygenic in that combined effects of variants in numerous
genes, plus nongenetic factors, determine outcome. The genes influencing common disease and quantitative traits remained largely unknown until the implementation in 2006 of genome-wide association
(GWA) studies that comprehensively surveyed common genetic variation (frequency >5%). By 2010, GWA studies identified >1,000 genetic
variants for polygenic traits. Typically, these variants together account
for a modest fraction (10%–30%) of heritability, but they have highlighted genes in both known and new biological pathways and genes
of unknown function. This initial effort prefigures new studies aimed
at rarer variation and decades of functional work to decipher newly
glimpsed biology. The greatest impact of GWA studies may not be in
predictive medicine but rather in the development over the next decades
of therapies based on novel biological insights.
11
ME62CH02-Hirschhorn
ARI
4 December 2010
11:3
DISEASE GENETICS:
SINGLE-GENE DISORDERS
AND POLYGENIC DISEASES
Annu. Rev. Med. 2011.62:11-24. Downloaded from www.annualreviews.org
by University of Oxford - Bodleian Library on 05/27/11. For personal use only.
Most diseases have an inherited component to
susceptibility, meaning that genetic variation
influences disease risk. The influence of genetics is clearest for Mendelian, single-gene disorders, where mutations in a particular gene
are necessary and sufficient to cause the disease. The transmission of these mutations from
parents to children typically results in clearly
recognizable familial patterns of inheritance,
because the penetrance of disease-causing mutations (the probability of being affected with
disease) is usually quite high, even though clinical presentation and severity can vary. These
highly penetrant mutations are usually evolutionarily deleterious and hence quite rare, and
many—but not all—are coding changes that
significantly disrupt or alter the protein encoded by the disease gene. Because of these
characteristics, genetic linkage analysis with
follow-up sequencing, which is particularly well
suited to finding highly penetrant mutations
with clear functional consequences, was highly
successful for mapping and identifying many
Mendelian disease genes (1, 2).
Most common diseases (for example, cancer, coronary heart disease, diabetes, and hypertension) and even quantitative traits (such
as height, weight, blood pressure, and blood
chemistries) also have an inherited component.
For most common diseases, heritability—the
fraction of variation in risk in a population that
is attributable to inherited factors—typically
ranges from 30% to well over 50% (for example, see 3–5). For quantitative traits, heritability estimates are often above 50% and
can approach 80%–90% (6–8). These diseases
and traits are polygenic, meaning that variation in many genes typically combine to influence phenotype (as do nongenetic factors
such as environmental exposures). Because no
single gene carries mutations that are sufficient to determine the outcome, families do
not display clearly recognizable patterns of
inheritance.
12
Hirschhorn
·
Gajdos
Given the success of linkage analysis in
Mendelian disorders, it was natural to attempt
linkage analysis for polygenic disorders.
However, with a few exceptions, large linkage
studies were unsuccessful at identifying regions
that met stringent statistical thresholds (2).
These important negative results suggested
that, for most polygenic diseases and quantitative traits, no individual gene contains
variants that together account for a substantial
fraction of heritability. Because linkage has low
power to detect genetic variants with modest
effects (9), especially if, as had been hypothesized, they are common in the population
(10–13), a different approach was proposed to
uncover the genetic basis of common disease:
association studies (see sidebar “Comparison
of Association and Linkage Analysis”).
GENETIC ASSOCIATION
STUDIES FOR POLYGENIC
DISEASES AND TRAITS
The initial association studies were candidate
gene studies, in which one or a handful of sequence variants from a single gene were tested
for correlation with disease. Associations with
human leukocyte antigen (HLA) were among
the first that could be readily reproduced, in
part because these often turned out to have
large effects on disease risk that could be reproducibly demonstrated even in small samples
of tens or hundreds of patients. For example,
nearly half of the excess risk to siblings of patients with type 1 diabetes is explained by variation at multiple genes within the HLA locus (see
14, 15 and references therein). A few early associations with type 1 diabetes and inflammatory
bowel disease (IBD) reflected effect sizes that
were large enough to produce signals in linkage analysis (14–16), and perhaps created expectations that many common diseases would
have variants with large effect sizes. Although
several common variants outside of the HLA
locus turned out to have moderate or large effects on diseases and quantitative traits (17–23),
most associated common variants have modest
Annu. Rev. Med. 2011.62:11-24. Downloaded from www.annualreviews.org
by University of Oxford - Bodleian Library on 05/27/11. For personal use only.
ME62CH02-Hirschhorn
ARI
4 December 2010
11:3
effects that require large association studies for
reliable detection (24, 25) and do not produce
readily apparent signals even in extremely large
linkage studies (2, 9).
The modest effect sizes of common variants
meant that early association studies, which
were limited by logistical considerations to
handfuls of variants tested in hundreds of samples, typically did not provide strong evidence
of association. Furthermore, nominally significant ( p < 0.05) results were often interpreted
as definitive, when in fact much lower p values
are required to demonstrate that an association
is likely valid (2, 25–27). Thus, false positive
and false negative results of association studies
were commonplace (24), and and the problem
was likely exacerbated by technical artifacts
and the presence of multiple genetic ancestries
within individual studies (population stratification) (28, 29). Gradually, well-powered and
well-controlled studies emerged that employed
large samples, methods to detect and correct
for population stratification (30), appropriate
statistical thresholds, and robust genotyping
methods. With these new standards in place,
association studies slowly began to yield
more reproducible results. With the development of genome-wide association (GWA)
studies, a more comprehensive approach
became available that could greatly accelerate
progress.
RESULTS OF THE FIRST WAVES
OF GENOME-WIDE
ASSOCIATION (GWA) STUDIES
The transition from candidate gene association
studies to GWA studies required four key technical and scientific advances: the discovery of
widespread correlation (linkage disequilibrium)
between groups of nearby common variants
(31); catalogues of common variants and their
correlations (32, 33); the development of highthroughput genotyping methods that could assay hundreds of thousands of genetic markers, typically single-nucleotide polymorphisms
(SNPs) (34); and statistical methods to use
the correlations between common variants to
COMPARISON OF ASSOCIATION
AND LINKAGE ANALYSIS
Association studies, like linkage studies, are designed to identify
regions of the genome that, when inherited, influence phenotype.
In a genetic association study of disease, DNA sequence variants
are genotyped in affected individuals and in controls (who can be
relatives or unrelated individuals of similar genetic ancestry). The
frequencies of genotypes (e.g., C/C versus C/T versus T/T) or
alleles (C versus T) are compared, and if one genotype or allele is
statistically significantly more frequent in cases than in controls,
the variant is said to be associated with the disease. Association
studies of quantitative traits are identical, except that variants are
tested for a correlation between genotype and trait values. The
fundamental difference between association and linkage studies
is that linkage analysis tests for correlation between a genomic
region and phenotype in families, whereas association analysis
searches for correlation in the population (essentially treating
the population as a single large family with an unknown pedigree). A major consequence of this difference is that association
analysis is better suited to finding common genetic variants with
more modest effects, whereas linkage analysis is better powered
to identify regions that contain variants (common or rare) that in
aggregate have a more substantial effect (see 2, 9). Furthermore,
because close relatives share stretches of genetic material over a
larger distance than do unrelated individuals, association analysis
can map disease-causing variants to a smaller region, defined by
the extent of correlation between variants in the population (linkage disequilibrium, or LD). Although these regions are smaller,
they often contain multiple genes; thus, association studies often
can highlight a small region but often cannot pinpoint a causal
gene within that region.
impute genotypes at additional variants not directly assayed by the genotyping platforms (35,
36). With the dramatic increase in scale from
candidate to GWA studies, increasing rigor and
sample sizes were essential to avoid a flood of
false positive and false negative results. Fortunately, the importance of stringent statistical
thresholds, high-quality data, and careful avoidance of artifacts were already becoming widely
accepted when GWA studies became feasible.
Consequently, most of the results from GWA
studies that surpass these stringent thresholds
of significance—“genome-wide significance,”
www.annualreviews.org • Genome-Wide Association Studies
13
ARI
4 December 2010
11:3
typically a p value threshold of 5 × 10−8 (25,
27)—have been replicable.
The first widely replicable result from
GWA studies was the association between variation at the complement factor H (CFH) gene
and age-related macular degeneration (AMD)
(18). This discovery was the earliest of its kind
in part because variation at CFH has a large
effect—greater than fourfold—on AMD risk.
Identification of associated variants for other
diseases and traits, where effect sizes are generally much more modest, awaited more robust
genotyping methods that could be applied in
large samples. The first such studies, published
in 2006 and the first half of 2007, typically
involved at most a few thousand individuals.
Although some of these did not identify any
clear associations, most identified one or a few
common genetic variants that were associated
at genome-wide levels of significance with
common diseases such as IBD (26, 37–39),
prostate cancer (40, 41), type 2 diabetes (26,
42–46), obesity (47), coronary heart disease
(26, 48), breast cancer (49, 50), and type 1 diabetes (26, 51). The GWA studies successfully
completed by the end of 2007 are too numerous
to list here, but these identified many more
novel associations between common variants
and polygenic diseases or quantitative traits.
These initial discoveries established several
themes. First, many of the associated common
genetic variants did not alter any predicted proteins, and some were hundreds of kilobases
away from the nearest gene, suggesting that
variation in the regulation of gene expression—
perhaps exerted over long distances—was likely
to be important in polygenic diseases and traits.
These results largely contrast with those of
single-gene disorders: Variants known to underlie single-gene disorders often are in the
coding region and severely alter protein function (52). Second, the individual effect sizes
of the variants from GWA studies were modest, typically explaining <1% of the population variation in phenotype. Thus, the initial
GWA studies did not substantially improve prediction of disease risk (53), again in contrast
to the experience from single-gene disorders.
Annu. Rev. Med. 2011.62:11-24. Downloaded from www.annualreviews.org
by University of Oxford - Bodleian Library on 05/27/11. For personal use only.
ME62CH02-Hirschhorn
14
Hirschhorn
·
Gajdos
Third, the GWA results were largely replicable in additional studies (25, 27), demonstrating the utility of combining careful study design with sufficient power to achieve rigorous
statistical thresholds, despite recent assertions
that many GWA results are artifactual (54). Finally, although some associated variants were
in or near known candidate genes, most associations highlighted genomic regions that contained no genes with previously known biological relevance, suggesting that these unbiased,
genome-wide genetic studies were providing
insights into novel biology (55). This pattern
is similar to the experience from single-gene
disorders, where the identification of diseasecausing genes often opened up new windows
into disease biology (2).
With the recognition that modest effect
sizes meant that large sample sizes were critical to achieve genome-wide statistical significance, the next two years saw the formation of
multiple consortia. Multiple groups with individual GWA studies joined forces to increase
sample size through meta-analysis of association results from each component study. These
efforts required a different, team-based scientific culture that recognized not only the efforts of key individuals but also the valuable
and substantial contributions of large numbers
(typically >100) of investigators. The result of
these collaborations was an explosion in new
associations. More than 1,000 genome-wide
significant associations between common genetic variants and polygenic traits and diseases
have been cataloged by Manolio’s group at the
National Human Genome Research Institute
(56; http://www.genome.gov/gwastudies/),
and most of these were published in 2009 and
2010.
The list of associations is too extensive and
too rapidly expanding to capture comprehensively here; instead, we briefly review the status (as of August 2010) of an illustrative group
of diseases and quantitative traits: IBD, type
2 diabetes, autism, lipid levels, height, and
hemoglobin F concentrations (Table 1). Given
the rate of discovery, the numbers of associations will no doubt continue to grow rapidly,
15
Abbreviations: N/A: Not applicable, as precise sites of action of therapies are not known or no therapies are available, or no known Mendelian forms of disease exist. UC: ulcerative colitis;
CD: Crohn’s disease.
Yes
Yes
∼35%
3
Hemoglobin F
levels
Height
www.annualreviews.org • Genome-Wide Association Studies
Identification of previously unsuspected key regulator
of globin switching, new potential drug targets
N/A
Additional pathways emerge after identification of
>100 loci; direct evidence for role of common
variants
Yes
Many
180
∼12%
Yes
Previously unsuspected genes validated as regulating
lipid levels, providing new potential drug targets
Yes
Near complete
95 across 3 traits
Lipid levels
1–2
Autism
∼25–30%
Yes
Rare variants, especially structural variants, more
prominent in initial genetic studies
No
N/A
∼10%
32
Type 2
diabetes
No
Yes
∼20%
>40
Inflammatory
bowel disease
Low
Yes
Comments
locus?
within loci?
N/A
(August 2010)
(August 2010)
Disease/trait
disorders
Multiple
variants per
Drug
targets
Overlap with
Mendelian
Heritability
explained
Like IBD, type 2 diabetes had been associated
with a few common variants prior to GWA
studies (see 63 and references therein). Nearly a
Number of loci
in GWA studies
Type 2 Diabetes
Characteristics of genome-wide association (GWA) discoveries for several exemplar common diseases and quantitative traits
Prior to GWA studies, several common or
moderately common variants had been associated with IBD. These were found by association studies that focused on genes within
regions highlighted by linkage studies or by
large-scale studies of nonsynonymous coding
variants (see 16, 57 and references therein).
Subsequent GWA studies identified 32 loci associated with Crohn’s disease (CD) and additional loci associated with ulcerative colitis
(UC), with partial overlap between the CD- and
UC-associated loci (see 16, 57 and references
therein, and 58–62).
The CD-associated loci account for ∼20%
of the heritable portion of disease risk, raising the theoretical possibility of predictive
and/or diagnostic genetic testing. However, the
greater impact of these studies has been to generate new understanding of the pathogenesis
of IBD. The associated loci define several biological pathways as causally involved in pathophysiology, including innate immunity, IL23
signaling/Th17 lymphocytes, autophagy, epithelial barrier integrity, and the secondary immune response (16, 57). Critically, several of
these pathways were not appreciated as being
involved in the pathogenesis of IBD until the
GWA results emerged. Some of these pathways
are active areas of research that may accelerate
the development of new therapies.
The comparison between the UC- and CDassociated loci has defined both common and
distinct pathways across these two related diseases (57), and the loci that are uniquely associated with UC or CD might eventually help
distinguish these two diseases, which in some
patients can be hard to distinguish clinically.
Table 1
Annu. Rev. Med. 2011.62:11-24. Downloaded from www.annualreviews.org
by University of Oxford - Bodleian Library on 05/27/11. For personal use only.
Inflammatory Bowel Disease
Yes
but the principles illustrated by these examples
are likely to remain relevant.
Cell cycle only statistically significant pathway, but
genes in loci suggest multiple known and novel
processes
11:3
Several clear, novel pathways guiding new therapeutic
research; variants might aid in distinguishing UC
versus CD
4 December 2010
Yes
ARI
No
ME62CH02-Hirschhorn
ARI
4 December 2010
11:3
dozen GWA studies identified a total of 32 loci
that together account for ∼10% of the heritable portion of disease risk (see 64 and references
therein). Although the fraction of variability in
disease risk explained by these variants is still
low, the variants can add information to known
risk factors, particularly earlier in life (65). The
variants are also beginning to outline known
and novel aspects of disease pathogenesis. Some
of the loci highlight known genes, such as those
encoding monogenic forms of diabetes (neonatal diabetes, mature-onset diabetes of youth, or
Wolfram syndrome), genes involved in insulin
signaling (IRS1), or genes encoding the sites
of action of antidiabetic medications (KCNJ11,
PPARG). Few pathways other than the cell cycle were unamibiguously defined by the genes
within associated loci, but some of the genes
suggest the involvement of less well-studied biological processes such as circadian rhythms,
beta cell zinc transport, and RNA binding proteins. These results suggest that many of the
pathways underlying the pathophysiology of
type 2 diabetes have yet to be defined or annotated. Examination of the effects of these variants in nondiabetic individuals demonstrated
that many associated variants increase diabetes
risk through diminished beta cell function, although some clearly affect insulin action.
Another interesting finding was that many of
the type 2 diabetes loci overlap with loci associated with other diseases or traits. In some cases
the associated variants are the same (and have
pleiotropic effects), but in many other cases
the associated variants differ, suggesting that
the variants exert different regulatory effects
on genes that play a role in the pathogenesis
of multiple diseases (64). Similar overlaps have
been seen with variants that influence the risk of
different autoimmune diseases (15) or different
types of cancer, most notably at several locations tens to hundreds of kilobases upstream of
the MYC gene (66). Finally, many of the variants that influence diabetes risk are quite distant from any known genes, raising the possibility of long-range regulation. Indeed, at the
MYC cancer-associated locus looping between
elements separated by hundreds of kilobases has
Annu. Rev. Med. 2011.62:11-24. Downloaded from www.annualreviews.org
by University of Oxford - Bodleian Library on 05/27/11. For personal use only.
ME62CH02-Hirschhorn
16
Hirschhorn
·
Gajdos
been documented, providing a possible mechanism for regulation over large distances (67).
Autism
Initial GWA studies of autism did not identify
any associated common variants, and only recently have two common variant associations
reached (or nearly reached) genome-wide levels of significance (68, 69). By contrast, several
studies have shown that rare structural variants (deletions or duplications) or point mutations can have substantial effects on disease risk
(70–72). The contrast between autism and most
other common diseases suggests that the genetic architecture of autism, and perhaps other
neuropsychiatric disorders such as schizophrenia (70, 71), may be somewhat different from
that of type 2 diabetes, IBD, and many other
common disorders.
Lipid Levels
Prior to the advent of GWA studies, several
genetic associations had already been identified with levels of HDL cholesterol, LDL
cholesterol, and triglycerides, including both
common and rare variants in known candidate
genes (73, 74). GWA studies have since
identified nearly 100 genetic loci with common
variants that together explain 25%–30% of the
population variation in these lipid levels (see
73, 75 and references therein). Almost every
gene known to underlie a single-gene disorder
of lipid metabolism has also been identified by
the GWA studies, not only demonstrating that
GWA studies are recapitulating the known
biology, but also suggesting that many of the
other genes identified by GWA studies will
be novel regulators of lipid metabolism and
have rare as well as common variants that influence lipid levels. Indeed, of the many genes that
are new candidates for playing a role in lipid
metabolism, several have recently been validated. By using in vitro and in vivo models, four
genes identified in GWA studies, PPP1R3B,
SORT1, TTC39B, and GALNT2, have been
confirmed as regulating human lipid levels
(75, 76). In the case of SORT1, there is
Annu. Rev. Med. 2011.62:11-24. Downloaded from www.annualreviews.org
by University of Oxford - Bodleian Library on 05/27/11. For personal use only.
ME62CH02-Hirschhorn
ARI
4 December 2010
11:3
compelling evidence that a particular regulatory variant is responsible for the variation
in LDL cholesterol levels (76). Two genes
encoding sites of action of lipid-lowering medications were highlighted in the GWA studies,
suggesting that some of the other genes with
no previously suspected connections to known
lipid metabolism pathways may provide suitable new drug targets. The variants discovered
by GWA studies of population-based samples
have also been examined for their ability to
predict clinical disease, and have been shown
to explain part of the risk for clinically defined
hypertriglyceridemia (77). Furthermore, some
of these variants contribute to the risk of
myocardial infarction, and, together with other
variants that are associated with coronary
heart disease, they provide nearly the same
discriminatory power for risk of coronary heart
disease as do individual traditional clinical risk
factors such as blood pressure (78). (We note
that when assessing predictive models, it is
important that the outcomes being predicted
align well with the phenotypes associated with
the variants used in the model.)
single GWA study, suggested that about half of
the heritability of height can be accounted for
by additive effects of a large number of common (>5% frequency) variants (81). At least 19
height-associated loci have multiple independently associated common variants, suggesting
that many of the as-yet-unidentified variants
may lie in already-identified loci. An enrichment of associated common missense variants
showed that, at least for some of the associated
loci, common variants themselves—rather than
collections of rare variants—were responsible
for the signals of association. Finally, the studies
of height also confirmed that, if enough loci are
identified, GWA studies can highlight biological pathways. For height, these include groups
of genes involved in chromatin structure,
hedgehog signaling, TGF-β signaling, growth
hormone/IGF-1 pathways, BMP/noggin signaling, and the extracellular matrix (80). Interestingly, some of these highly relevant pathways
emerged only after hundreds of associations
were identified, suggesting that very large numbers of loci may have a greater impact on implicating biological processes than on explaining
a larger fraction of phenotypic variation.
Height
Height is the classic polygenic trait (79), is
highly heritable (8), and is routinely measured
in large populations. Thus, quite large GWA
studies of height are feasible and may provide
early general insights into the genetic architecture of complex traits. Prior to GWA studies, no common variants were known to be
associated with height, but a series of increasingly large GWA studies has now been completed. The most recent involved a first stage
of 129,000 individuals and a replication stage
of up to ∼50,000 individuals; it identified variants at 180 loci that together explain ∼12%
of the heritable variation in height (80). Power
considerations suggested that there are several
hundred more common variants yet to be discovered that have similar effect sizes on height,
and that together these common variants will
explain at least ∼2% of heritability. A separate analysis, using data from all the SNPs in a
Hemoglobin F Levels
During human development, globin gene expression shifts from the fetal form (hemoglobin
F, HbF) to the adult form (hemoglobin A), but
detectable levels of HbF can still be found in
adults due to residual expression of the gamma
globin gene. HbF levels are an important modifier of the severity of sickle-cell disease, thalassemia, and other hemoglobinopathies, and
are under genetic regulation, partially explained
by rare and common cis-acting variation at the
globin locus (21). However, despite decades of
intensive efforts, the genes responsible for regulating the switch from gamma to beta globin expression were not known. The first GWA studies identified common variants that accounted
for >30% of the population variability in HbF
levels, including variation at the globin locus
and multiple variants at two other loci, near
BCL11A and MYB-HBS1L (19, 20, 22). These
www.annualreviews.org • Genome-Wide Association Studies
17
ME62CH02-Hirschhorn
ARI
4 December 2010
11:3
variants can account for some of the variability
in severity of sickle-cell disease or thalassemia
(22, 82), but the major impact of these discoveries is likely to be through the new understanding of globin regulation. Specifically, the
BCL11A protein was rapidly shown to be a key
regulator of globin switching in vitro and in vivo
(83, 84). Thus, GWA studies have opened new
areas of research that could plausibly lead to
new therapies targeting these proteins.
Annu. Rev. Med. 2011.62:11-24. Downloaded from www.annualreviews.org
by University of Oxford - Bodleian Library on 05/27/11. For personal use only.
THE NEXT WAVE OF GENETIC
STUDIES OF POLYGENIC TRAITS
AND DISEASES
The first three years’ worth of GWA studies
discovered numerous associated genetic loci for
many different polygenic traits and diseases.
However, the variants identified thus far generally explain less than half (and in a number
of cases less than 5%) of the heritability. Several possible explanations of this “missing heritability” have been proposed (81, 85, 86). These
include a role for additional common variants
of small effect, a role for less common variants
(including variants with frequencies between
0.5% and 5%, which have not been well surveyed by current GWA studies), the presence
of even rarer variants (those that are essentially
private to each unrelated individual), nonadditive interactions between variants, and epigenetic effects. Regardless, these initial GWA
studies, along with the failure of linkage studies to generate reproducible signals, have provided strong evidence that many common diseases and quantitative traits are quite polygenic,
and this is likely an important contributor to asyet-unexplained variance.
Indeed, despite the large sizes of GWA studies to date, it remains possible that much of the
unexplained variance can be accounted for by
additional common variants that have not yet
been detected. Recent work suggests that common variants of small effect may account for up
to half of the heritability of phenotypes such
as height and even schizophrenia (80, 81, 87),
so further increases in sample size will likely
yield new associated variants. Current GWA
18
Hirschhorn
·
Gajdos
studies have largely interrogated variants with
frequencies of at least 5%–10% in the population. The next versions of GWA studies could
build on data from the 1000 Genomes Project
(http://www.1000genomes.org) to identify
additional associated variants, as exemplified by
a recent candidate gene sequencing study of
type 1 diabetes that found associated variants
in the 1% frequency range in IFIH1 (88).
Advances in sequencing technologies will
enable genome-wide sequencing studies to
assess more comprehensively the role of rare
variation. These will no doubt identify rare
variation that influences polygenic traits and
diseases, based on early successes of classic
candidate gene sequencing studies of obesity,
lipid levels, and blood pressure (74, 89–91).
However, implementation and interpretation
of sequencing studies in polygenic traits and
diseases will not be straightforward. First,
methods for rigorous analysis, high-quality
data generation, careful interpretation, and
avoidance of artifact (especially confounding
by ancestry) have not been well-established
for sequencing studies. Second, in sequencing
studies, rare variants must be analyzed in
aggregate, and it may be difficult to distinguish
rare functional variants from the background
of rare neutral variants, particularly for noncoding variation. Finally, the lack of strong
linkage signals provides evidence that the
contribution of rare variants at any one gene
will be modest, so it is likely that large samples
will be required, perhaps comparable to those
that have been needed for GWA studies (2).
IMPACT OF GWA STUDIES ON
CLINICAL MEDICINE
The identification of genetic loci can have two
major impacts on clinical medicine—through
prediction of future outcomes or through elucidation of underlying biology (Figure 1). Prediction has the potential for immediate impact,
although, as discussed below and by others (53,
92), the clinical utility of predictive genetic tests
depends both on the degree of additional prediction provided by the genetic markers and on
ME62CH02-Hirschhorn
ARI
4 December 2010
11:3
whether the clinical course of action would be
altered by predictive information. By contrast,
the insights into biology and pathophysiology
are likely to develop over a longer time frame.
New genetic
loci discovered
Annu. Rev. Med. 2011.62:11-24. Downloaded from www.annualreviews.org
by University of Oxford - Bodleian Library on 05/27/11. For personal use only.
Prediction Using Genetics: The
Potential for “Personalized Medicine”
Much of the immediate focus on the discoveries
from GWA studies has been on the feasibility
of using genetic testing to predict disease or to
guide therapy. One illustration is the rapid appearance of direct-to-consumer genetic testing
companies that utilize common variants to provide estimates of relative risk, but not absolute
risk (93). However, it is not certain that current
or even future collections of associated variants
will be clinically useful for prediction. First, the
level of prediction thus far is typically modest, although in some cases the discriminative
power of common genotypes is comparable to
clinical tests that are in widespread use [such as
LDL cholesterol for the prediction of myocardial infarction (78)]. More critically, the clinical utility of predictive models depends on their
ability to alter clinical management, and this in
turn depends on the availability of clinical interventions that are not universally indicated for
reasons of safety or cost, but that might on balance be beneficial or cost-effective for higherrisk subgroups of patients (Figure 1).
There are many examples of genetics influencing clinical management for Mendelian or
near-Mendelian traits and diseases, including
diagnosis, prognosis, and reproductive counseling for many single-gene disorders, as well as
blood typing prior to transfusion. In a few polygenic cases, particularly those involving drug
response, the predictive information is strong
and is used to make management decisions; for
example, patients are HLA-typed to avoid hypersensitivity reactions to abacavir (94). Other
GWA results in the area of pharmacogenetics
may lead to similar clinical algorithms that incorporate genetic testing, perhaps with genetically guided interventions to avoid adverse reactions (95). In other cases, predictive power
is strong but there are no preventive measures
Short
time frame
Short-to-medium
time frame
Identification of
the relevant
variants
Identification of
the relevant
genes
Short
time frame
Prediction
models
Medium-to-long
time frame
Understanding
of underlying
biology
Depends on predictive
power and available
clinical options
Clinically useful
prediction
Long
time frame
Clinically useful
therapies
Figure 1
Identification of genetic loci that are associated with human disease can lead in
two directions. Left: In the short term, if the predictive value of specific genetic
variants (which can be nearby markers of the actual causal variants) is strong
enough, and there are clinical interventions whose utility depends on predictive
information, genetic information can guide clinical care. The clearest examples
come from single-gene disorders and pharmacogenetics, where predictive
power is high. However, many clinically useful tests have predictive power
similar to that provided by genetic variation for many common diseases, raising
the possibility of eventual application of genetic information for prediction.
Right: Alternatively, identification of genetic loci and the relevant genes at
those loci can lead to a deeper understanding of the underlying biology, which
can in turn help guide the development of new interventions and therapies.
Even when successful, this process usually takes decades before new therapies
are developed, but it may provide the greater long-term impact of human
genetic studies.
that are not universally indicated, so the associations do not affect clinical management. Examples include type 1 diabetes and AMD. For
yet other diseases, the level of prediction may
be lower, but might still be sufficient to identify
a subgroup of patients who would benefit from
an intervention that is either too expensive or
too risky to apply to the general population.
One possible example is the use of intensive
lifestyle management and/or pharmacotherapy
in patients who are at high risk of developing type 2 diabetes; although these interventions are effective (96), they are too costly for
www.annualreviews.org • Genome-Wide Association Studies
19
ME62CH02-Hirschhorn
ARI
4 December 2010
11:3
Annu. Rev. Med. 2011.62:11-24. Downloaded from www.annualreviews.org
by University of Oxford - Bodleian Library on 05/27/11. For personal use only.
universal adoption, but genetic test results in
combination with other known risk factors
might eventually identify a high-risk group
for whom the interventions would be costeffective. As predictive power continues to improve, clinical trials that test whether it is useful
to take genetic factors into consideration could
become more informative. Finally, we note
that genetic risk scores would most appropriately supplement rather than replace traditional
risk estimates, and should be evaluated in this
context.
Understanding Human Biology
and Pathophysiology
Looking into the future, the greatest impact of
GWA studies (and genetic studies more generally) will likely flow from expanding our understanding of underlying biology in patients.
GWA studies of polygenic diseases and traits
recapitulate many known biological pathways
relevant to those diseases and traits. Furthermore, the delineation of pathways becomes
clearer as the number of loci increases, with the
picture continuing to sharpen even when the
number of loci reaches the hundreds (80), so
the continued identification of loci may aid in
the identification of new pathways. Even more
important, the sites of action of many known
therapies have already been highlighted by associated variants (55, 75, 80), suggesting that
the other loci may provide clues to new therapies. It is clear that many GWA loci do not contain genes that had previously been suspected
as biological candidates. In some cases (as described in this review), these new genes identify well-studied biological pathways that were
not previously connected to a particular disease,
such as complement factor H and AMD, or autophagy and Crohn’s disease. In these cases, efforts directed toward developing potential therapies could begin immediately. In others, more
work is needed to understand the clues that have
been provided by GWA studies.
20
Hirschhorn
·
Gajdos
It is important to realize that, because of the
genetic correlation (linkage disequilibrium)
between nearby variants (see sidebar “Comparison of Association and Linkage Analysis”), association studies do not usually pinpoint the variant responsible for the association, or even the
gene that is affected by the associated variant.
Instead, association studies typically highlight
a small region that may contain several genes
(although sometimes the relevant gene may
be obvious from previous genetic or biological
studies). Although many GWA studies typically
report the nearest gene(s) as a way of identifying
the location of an associated variant, connecting
a specific gene with a trait or disease generally
requires additional genetic, biological, or
functional data. Thus, additional functional or
genetic studies (e.g., sequencing) will often be
needed in order to identify the relevant gene
as a prerequisite to deciphering the relevant
biology.
Even if the relevant genes and biological
pathways are understood, the time frame between biological knowledge and a new therapy
can be long, as illustrated by the two decades
that elapsed between Goldstein & Brown’s
work implicating HMGCR and cholesterol
metabolism in atherosclerosis (97) and the successful use of HMGCR inhibitors (statins) to
prevent coronary artery disease (98). Where the
genes and/or biology are unknown, the time
frame will be longer. Still, GWA studies (and
the studies that will follow in their wake) hold
the promise of new and better therapies that
target previously unsuspected aspects of disease
biology that are genetically validated as relevant in patients. The promise of revolutionizing clinical medicine is great enough to justify
follow-up studies based on the findings from
these genetic studies. However, the likely time
frame of these follow-up studies is long enough
that the true level of success of genetic studies
of common diseases and polygenic traits will be
best judged in the coming decades rather than
in the next few months or years.
ME62CH02-Hirschhorn
ARI
4 December 2010
11:3
DISCLOSURE STATEMENT
The authors are not aware of any affiliations, memberships, funding, or financial holdings that
might be perceived as affecting the objectivity of this review.
Annu. Rev. Med. 2011.62:11-24. Downloaded from www.annualreviews.org
by University of Oxford - Bodleian Library on 05/27/11. For personal use only.
LITERATURE CITED
1. Glazier AM, Nadeau JH, Aitman TJ. 2002. Finding genes that underlie complex traits. Science 298:2345–
49
2. Altshuler D, Daly MJ, Lander ES. 2008. Genetic mapping in human disease. Science 322:881–88
3. Kaprio J, Tuomilehto J, Koskenvuo M, et al. 1992. Concordance for type 1 (insulin-dependent) and type
2 (non-insulin-dependent) diabetes mellitus in a population-based cohort of twins in Finland. Diabetologia
35:1060–67
4. Zdravkovic S, Wienke A, Pedersen NL, et al. 2002. Heritability of death from coronary heart disease: a
36-year follow-up of 20 966 Swedish twins. J. Intern. Med. 252:247–54
5. Sullivan PF, Kendler KS, Neale MC. 2003. Schizophrenia as a complex trait: evidence from a meta-analysis
of twin studies. Arch. Gen. Psychiatry 60:1187–92
6. Beekman M, Heijmans BT, Martin NG, et al. 2002. Heritabilities of apolipoprotein and lipid levels in
three countries. Twin Res. 5:87–97
7. Maes HH, Neale MC, Eaves LJ. 1997. Genetic and environmental factors in relative body weight and
human adiposity. Behav. Genet. 27:325–51
8. Silventoinen K, Sammalisto S, Perola M, et al. 2003. Heritability of adult body height: a comparative
study of twin cohorts in eight countries. Twin Res. 6:399–408
9. Risch N, Merikangas K. 1996. The future of genetic studies of complex human diseases. Science 273:1516–
17
10. Collins FS, Guyer MS, Charkravarti A. 1997. Variations on a theme: cataloging human DNA sequence
variation. Science 278:1580–81
11. Harris H. 1975. The Principles of Human Biochemical Genetics. Amsterdam: North-Holland
12. Lander ES. 1996. The new genomics: global views of biology. Science 274:536–39
13. Reich DE, Lander ES. 2001. On the allelic spectrum of human disease. Trends Genet. 17:502–10
14. Clayton DG. 2009. Prediction and interaction in complex disease genetics: experience in type 1 diabetes.
PLoS Genet. 5:e1000540
15. Zhernakova A, van Diemen CC, Wijmenga C. 2009. Detecting shared pathogenesis from the shared
genetics of immune-related diseases. Nat. Rev. Genet. 10:43–55
16. Van Limbergen J, Wilson DC, Satsangi J. 2009. The genetics of Crohn’s disease. Annu. Rev. Genomics
Hum. Genet. 10:89–116
17. Poirier J, Davignon J, Bouthillier D, et al. 1993. Apolipoprotein E polymorphism and Alzheimer’s disease.
Lancet 342:697–99
18. Klein RJ, Zeiss C, Chew EY, et al. 2005. Complement factor H polymorphism in age-related macular
degeneration. Science 308:385–89
19. Thein SL, Menzel S, Peng X, et al. 2007. Intergenic variants of HBS1L-MYB are responsible for a major
quantitative trait locus on chromosome 6q23 influencing fetal hemoglobin levels in adults. Proc. Natl.
Acad. Sci. USA 104:11346–51
20. Menzel S, Garner C, Gut I, et al. 2007. A QTL influencing F cell production maps to a gene encoding a
zinc-finger protein on chromosome 2p15. Nat. Genet. 39:1197–99
21. Thein SL, Menzel S, Lathrop M, et al. 2009. Control of fetal hemoglobin: new insights emerging from
genomics and clinical implications. Hum. Mol. Genet. 18:R216–23
22. Uda M, Galanello R, Sanna S, et al. 2008. Genome-wide association study shows BCL11A associated with
persistent fetal hemoglobin and amelioration of the phenotype of beta-thalassemia. Proc. Natl. Acad. Sci.
USA 105:1620–25
23. Jonas DE, McLeod HL. 2009. Genetic and clinical factors relating to warfarin dosing. Trends Pharmacol.
Sci. 30:375–86
www.annualreviews.org • Genome-Wide Association Studies
21
ARI
4 December 2010
11:3
24. Lohmueller KE, Pearce CL, Pike M, et al. 2003. Meta-analysis of genetic association studies supports a
contribution of common variants to susceptibility to common disease. Nat. Genet. 33:177–82
25. McCarthy MI, Abecasis GR, Cardon LR, et al. 2008. Genome-wide association studies for complex traits:
consensus, uncertainty and challenges. Nat. Rev. Genet. 9:356–69
26. Wellcome Trust Case Control Consortium. 2007. Genome-wide association study of 14,000 cases of
seven common diseases and 3,000 shared controls. Nature 447:661–78
27. Chanock SJ, Manolio T, Boehnke M, et al. 2007. Replicating genotype-phenotype associations. Nature
447:655–60
28. Clayton DG, Walker NM, Smyth DJ, et al. 2005. Population structure, differential bias and genomic
control in a large-scale, case-control association study. Nat. Genet. 37:1243–46
29. Campbell CD, Ogburn EL, Lunetta KL, et al. 2005. Demonstrating stratification in a European American
population. Nat. Genet. 37:868–72
30. Price AL, Zaitlen NA, Reich D, et al. 2010. New approaches to population stratification in genome-wide
association studies. Nat. Rev. Genet. 11:459–63
31. Gabriel SB, Schaffner SF, Nguyen H, et al. 2002. The structure of haplotype blocks in the human genome.
Science 296:2225–29
32. International HapMap Consortium. 2005. A haplotype map of the human genome. Nature 437:1299–320
33. Frazer KA, Ballinger DG, Cox DR, et al. 2007. A second generation human haplotype map of over
3.1 million SNPs. Nature 449:851–61
34. Ragoussis J. 2009. Genotyping technologies for genetic research. Annu. Rev. Genomics Hum. Genet. 10:117–
33
35. Marchini J, Howie B. 2010. Genotype imputation for genome-wide association studies. Nat. Rev. Genet.
11:499–511
36. Li Y, Willer C, Sanna S, et al. 2009. Genotype imputation. Annu. Rev. Genomics Hum. Genet. 10:387–406
37. Duerr RH, Taylor KD, Brant SR, et al. 2006. A genome-wide association study identifies IL23R as an
inflammatory bowel disease gene. Science 314:1461–63
38. Rioux JD, Xavier RJ, Taylor KD, et al. 2007. Genome-wide association study identifies new susceptibility
loci for Crohn disease and implicates autophagy in disease pathogenesis. Nat. Genet. 39:596–604
39. Parkes M, Barrett JC, Prescott NJ, et al. 2007. Sequence variants in the autophagy gene IRGM and
multiple other replicating loci contribute to Crohn’s disease susceptibility. Nat. Genet. 39:830–32
40. Amundsdottir LT, Sulem P, Gudmundsson J, et al. 2006. A common variant associated with prostate
cancer in European and African populations. Nat. Genet. 38:652–58
41. Gudmundsson J, Sulem P, Manolescu A, et al. 2007. Genome-wide association study identifies a second
prostate cancer susceptibility variant at 8q24. Nat. Genet. 39:631–37
42. Sladek R, Rocheleau G, Rung J, et al. 2007. A genome-wide association study identifies novel risk loci for
type 2 diabetes. Nature 445:881–85
43. Steinthorsdottir V, Thorleifsson G, Reynisdottir I, et al. 2007. A variant in CDKAL1 influences insulin
response and risk of type 2 diabetes. Nat. Genet. 39:770–75
44. Saxena R, Voight BF, Lyssenko V, et al. 2007. Genome-wide association analysis identifies loci for type
2 diabetes and triglyceride levels. Science 316:1331–36
45. Scott LJ, Mohlke KL, Bonnycastle LL, et al. 2007. A genome-wide association study of type 2 diabetes
in Finns detects multiple susceptibility variants. Science 316:1341–45
46. Zeggini E, Weedon MN, Lindgren CM, et al. 2007. Replication of genome-wide association signals in
UK samples reveals risk loci for type 2 diabetes. Science 316:1336–41
47. Frayling TM, Timpson NJ, Weedon MN, et al. 2007. A common variant in the FTO gene is associated
with body mass index and predisposes to childhood and adult obesity. Science 316:889–94
48. McPherson R, Pertsemlidis A, Kavaslar N, et al. 2007. A common allele on chromosome 9 associated with
coronary heart disease. Science 316:1488–91
49. Easton DF, Pooley KA, Dunning AM, et al. 2007. Genome-wide association study identifies novel breast
cancer susceptibility loci. Nature 447:1087–93
50. Hunter DJ, Kraft P, Jacobs KB, et al. 2007. A genome-wide association study identifies alleles in FGFR2
associated with risk of sporadic postmenopausal breast cancer. Nat. Genet. 39:870–74
Annu. Rev. Med. 2011.62:11-24. Downloaded from www.annualreviews.org
by University of Oxford - Bodleian Library on 05/27/11. For personal use only.
ME62CH02-Hirschhorn
22
Hirschhorn
·
Gajdos
Annu. Rev. Med. 2011.62:11-24. Downloaded from www.annualreviews.org
by University of Oxford - Bodleian Library on 05/27/11. For personal use only.
ME62CH02-Hirschhorn
ARI
4 December 2010
11:3
51. Todd JA, Walker NM, Cooper JD, et al. 2007. Robust associations of four new chromosome regions from
genome-wide analyses of type 1 diabetes. Nat. Genet. 39:857–64
52. Botstein D, Risch N. 2003. Discovering genotypes underlying human phenotypes: past successes for
Mendelian disease, future approaches for complex disease. Nat. Genet. 33(Suppl.):228–37
53. Kraft P, Hunter DJ. 2009. Genetic risk prediction—Are we there yet? N. Engl. J. Med. 360:1701–3
54. McClellan J, King MC. 2010. Genetic heterogeneity in human disease. Cell 141:210–17
55. Hirschhorn JN. 2009. Genomewide association studies—illuminating biologic pathways. N. Engl. J. Med.
360:1699–701
56. Hindorff LA, Sethupathy P, Junkins HA, et al. 2009. Potential etiologic and functional implications of
genome-wide association loci for human diseases and traits. Proc. Natl. Acad. Sci. USA 106:9362–67
57. Rosenstiel P, Sina C, Franke A, et al. 2009. Towards a molecular risk map—recent advances on the
etiology of inflammatory bowel disease. Semin. Immunol. 21:334–45
58. Barrett JC, Lee JC, Lees CW, et al. 2009. Genome-wide association study of ulcerative colitis identifies
three new susceptibility loci, including the HNF4A region. Nat. Genet. 41:1330–34
59. Asano K, Matsushita T, Umeno J, et al. 2009. A genome-wide association study identifies three new
susceptibility loci for ulcerative colitis in the Japanese population. Nat. Genet. 41:1325–29
60. Imielinski M, Baldassano RN, Griffiths A, et al. 2009. Common variants at five new loci associated with
early-onset inflammatory bowel disease. Nat. Genet. 41:1335–40
61. Franke A, Balschun T, Sina C, et al. 2010. Genome-wide association study for ulcerative colitis identifies
risk loci at 7q22 and 22q13 (IL17REL). Nat. Genet. 42:292–94
62. McGovern DP, Gardet A, Torkvist L, et al. 2010. Genome-wide association identifies multiple ulcerative
colitis susceptibility loci. Nat. Genet. 42:332–37
63. Florez JC, Hirschhorn J, Altshuler D. 2003. The inherited basis of diabetes mellitus: implications for the
genetic analysis of complex traits. Annu. Rev. Genomics Hum. Genet. 4:257–91
64. Voight BF, Scott LJ, Steinthorsdottir V, et al. 2010. Twelve type 2 diabetes susceptibility loci identified
through large-scale association analysis. Nat. Genet. 42:579–89
65. Lyssenko V, Jonsson A, Almgren P, et al. 2008. Clinical risk factors, DNA variants, and the development
of type 2 diabetes. N. Engl. J. Med. 359:2220–32
66. Ghoussaini M, Song H, Koessler T, et al. 2008. Multiple loci with different cancer specificities within the
8q24 gene desert. J. Natl. Cancer Inst. 100:962–66
67. Ahmadiyeh N, Pomerantz MM, Grisanzio C, et al. 2010. 8q24 prostate, breast, and colon cancer risk loci
show tissue-specific long-range interaction with MYC. Proc. Natl. Acad. Sci. USA 107:9742–46
68. Wang K, Zhang H, Ma D, et al. 2009. Common genetic variants on 5p14.1 associate with autism spectrum
disorders. Nature 459:528–33
69. Weiss LA, Arking DE, Daly MJ, et al. 2009. A genome-wide linkage and association scan reveals novel
loci for autism. Nature 461:802–8
70. Merikangas AK, Corvin AP, Gallagher L. 2009. Copy-number variants in neurodevelopmental disorders:
promises and challenges. Trends Genet. 25:536–44
71. Stankiewicz P, Lupski JR. 2010. Structural variation in the human genome and its role in disease. Annu.
Rev. Med. 61:437–55
72. Berkel S, Marshall CR, Weiss B, et al. 2010. Mutations in the SHANK2 synaptic scaffolding gene in
autism spectrum disorder and mental retardation. Nat. Genet. 42:489–91
73. Hegele RA. 2009. Plasma lipoproteins: genetic influences and clinical implications. Nat. Rev. Genet.
10:109–21
74. Cohen JC, Kiss RS, Pertsemlidis A, et al. 2004. Multiple rare alleles contribute to low plasma levels of
HDL cholesterol. Science 305:869–72
75. Teslovich TM, Musunuru K, Smith AV, et al. 2010. Biological, clinical and population relevance of 95
loci for blood lipids. Nature 466:707–13
76. Musunuru K, Strong A, Frank-Kamenetsky M, et al. 2010. From noncoding variant to phenotype via
SORT1 at the 1p13 cholesterol locus. Nature 466:714–19
77. Johansen CT, Wang J, Lanktree MB, et al. 2010. Excess of rare variants in genes identified by genome-wide
association study of hypertriglyceridemia. Nat. Genet. 42:684–87
www.annualreviews.org • Genome-Wide Association Studies
23
ARI
4 December 2010
11:3
78. Ripatti S, Tikkanen E, Orho-Melander M, et al. 2010. A multi-locus genetic risk score and risk for incident
coronary heart disease. Lancet 376:1393–400
79. Fisher RA. 1918. The correlation between relatives on the supposition of Mendelian inheritance. Trans.
R. Soc. Edinb. 52:399–433
80. Lango Allen H, Estrada K, Lettre G, et al. 2010. Hundreds of variants influence height and cluster in
genomic loci and biological pathways. Nature 467:832–38
81. Yang J, Benyamin B, McEvoy BP, et al. 2010. Common SNPs explain a large proportion of the heritability
for human height. Nat. Genet. 42:565–69
82. Lettre G, Sankaran VG, Bezerra MA, et al. 2008. DNA polymorphisms at the BCL11A, HBS1L-MYB,
and beta-globin loci associate with fetal hemoglobin levels and pain crises in sickle cell disease. Proc. Natl.
Acad. Sci. USA 105:11869–74
83. Sankaran VG, Xu J, Ragoczy T, et al. 2009. Developmental and species-divergent globin switching are
driven by BCL11A. Nature 460:1093–97
84. Sankaran VG, Menne TF, Xu J, et al. 2008. Human fetal hemoglobin expression is regulated by the
developmental stage-specific repressor BCL11A. Science 322:1839–42
85. Manolio TA, Collins FS, Cox NJ, et al. 2009. Finding the missing heritability of complex diseases. Nature
461:747–53
86. McCarthy MI, Hirschhorn JN. 2008. Genome-wide association studies: potential next steps on a genetic
journey. Hum. Mol. Genet. 17:R156–65
87. Purcell SM, Wray NR, Stone JL, et al. 2009. Common polygenic variation contributes to risk of
schizophrenia and bipolar disorder. Nature 460:748–52
88. Nejentsev S, Walker N, Riches D, et al. 2009. Rare variants of IFIH1, a gene implicated in antiviral
responses, protect against type 1 diabetes. Science 324:387–89
89. Farooqi IS, Yeo GS, Keogh JM, et al. 2000. Dominant and recessive inheritance of morbid obesity
associated with melanocortin 4 receptor deficiency. J. Clin. Invest. 106:271–79
90. Ji W, Foo JN, O’Roak BJ, et al. 2008. Rare independent mutations in renal salt handling genes contribute
to blood pressure variation. Nat. Genet. 40:592–99
91. Vaisse C, Clement K, Durand E, et al. 2000. Melanocortin-4 receptor mutations are a frequent and
heterogeneous cause of morbid obesity. J. Clin. Invest. 106:253–62
92. Rogowski WH, Grosse SD, Khoury MJ. 2009. Challenges of translating genetic tests into clinical and
public health practice. Nat. Rev. Genet. 10:489–95
93. Kaye J. 2008. The regulation of direct-to-consumer genetic tests. Hum. Mol. Genet. 17:R180–83
94. Lai-Goldman M, Faruki H. 2008. Abacavir hypersensitivity: a model system for pharmacogenetic test
adoption. Genet. Med. 10:874–78
95. Fellay J, Thompson AJ, Ge D, et al. 2010. ITPA gene variants protect against anaemia in patients treated
for chronic hepatitis C. Nature 464:405–8
96. Knowler WC, Barrett-Connor E, Fowler SE, et al. 2002. Reduction in the incidence of type 2 diabetes
with lifestyle intervention or metformin. N. Engl. J. Med. 346:393–403
97. Goldstein JL, Brown MS. 1973. Familial hypercholesterolemia: identification of a defect in the regulation of 3-hydroxy-3-methylglutaryl coenzyme A reductase activity associated with overproduction of
cholesterol. Proc. Natl. Acad. Sci. USA 70:2804–8
98. Pedersen TR, Kjekshus J, Berg K, et al. 1994. Randomised trial of cholesterol lowering in 4444 patients
with coronary heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet 344:1383–89
Annu. Rev. Med. 2011.62:11-24. Downloaded from www.annualreviews.org
by University of Oxford - Bodleian Library on 05/27/11. For personal use only.
ME62CH02-Hirschhorn
24
Hirschhorn
·
Gajdos
ME62-Frontmatter
ARI
15 December 2010
15:56
Annual Review of
Medicine
Annu. Rev. Med. 2011.62:11-24. Downloaded from www.annualreviews.org
by University of Oxford - Bodleian Library on 05/27/11. For personal use only.
Contents
Volume 62, 2011
Role of Postmarketing Surveillance in Contemporary Medicine
Janet Woodcock, Rachel E. Behrman, and Gerald J. Dal Pan p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 1
Genome-Wide Association Studies: Results from the First Few Years
and Potential Implications for Clinical Medicine
Joel N. Hirschhorn and Zofia K.Z. Gajdos p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p11
Imaging of Atherosclerosis
D.R.J. Owen, A.C. Lindsay, R.P. Choudhury, and Z.A. Fayad p p p p p p p p p p p p p p p p p p p p p p p p p p p25
Novel Oral Factor Xa and Thrombin Inhibitors in the Management
of Thromboembolism
Bengt I. Eriksson, Daniel J. Quinlan, and John W. Eikelboom p p p p p p p p p p p p p p p p p p p p p p p p p p p41
The Fabry Cardiomyopathy: Models for the Cardiologist
Frank Weidemann, Markus Niemann, David G. Warnock, Georg Ertl,
and Christoph Wanner p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p59
Kawasaki Disease: Novel Insights into Etiology
and Genetic Susceptibility
Anne H. Rowley p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p69
State of the Art in Therapeutic Hypothermia
Joshua W. Lampe and Lance B. Becker p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p79
Therapeutic Potential of Lung Epithelial Progenitor Cells Derived
from Embryonic and Induced Pluripotent Stem Cells
Rick A. Wetsel, Dachun Wang, and Daniel G. Calame p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p95
Therapeutics Development for Cystic Fibrosis: A Successful Model
for a Multisystem Genetic Disease
Melissa A. Ashlock and Eric R. Olson p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 107
Early Events in Sexual Transmission of HIV and SIV
and Opportunities for Interventions
Ashley T. Haase p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 127
HIV Infection, Inflammation, Immunosenescence, and Aging
Steven G. Deeks p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 141
v
ME62-Frontmatter
ARI
15 December 2010
15:56
The Increasing Burden of HIV-Associated Malignancies
in Resource-Limited Regions
Corey Casper p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 157
Biliary Atresia: Will Blocking Inflammation Tame the Disease?
Kazuhiko Bessho and Jorge A. Bezerra p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 171
Advances in Palliative Medicine and End-of-Life Care
Janet L. Abrahm p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 187
Annu. Rev. Med. 2011.62:11-24. Downloaded from www.annualreviews.org
by University of Oxford - Bodleian Library on 05/27/11. For personal use only.
Clostridium difficile and Methicillin-Resistant Staphylococcus aureus:
Emerging Concepts in Vaccine Development
David C. Kaslow and John W. Shiver p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 201
Antiestrogens and Their Therapeutic Applications in Breast Cancer
and Other Diseases
Simak Ali, Laki Buluwela, and R. Charles Coombes p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 217
Mechanisms of Endocrine Resistance in Breast Cancer
C. Kent Osborne and Rachel Schiff p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 233
Multiple Myeloma
Jacob Laubach, Paul Richardson, and Kenneth Anderson p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 249
Muscle Wasting in Cancer Cachexia: Clinical Implications, Diagnosis,
and Emerging Treatment Strategies
Shontelle Dodson, Vickie E. Baracos, Aminah Jatoi, William J. Evans,
David Cella, James T. Dalton, and Mitchell S. Steiner p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 265
Pharmacogenetics of Endocrine Therapy for Breast Cancer
Michaela J. Higgins and Vered Stearns p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 281
Therapeutic Approaches for Women Predisposed to Breast Cancer
Katherine L. Nathanson and Susan M. Domchek p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 295
New Approaches to the Treatment of Osteoporosis
Barbara C. Silva and John P. Bilezikian p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 307
Regulation of Bone Mass by Serotonin: Molecular Biology
and Therapeutic Implications
Gerard Karsenty and Vijay K. Yadav p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 323
Alpha-1-Antitrypsin Deficiency: Importance of Proteasomal
and Autophagic Degradative Pathways in Disposal of Liver
Disease–Associated Protein Aggregates
David H. Perlmutter p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 333
Hepcidin and Disorders of Iron Metabolism
Tomas Ganz and Elizabeta Nemeth p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 347
vi
Contents
ME62-Frontmatter
ARI
15 December 2010
15:56
Interactions Between Gut Microbiota and Host Metabolism
Predisposing to Obesity and Diabetes
Giovanni Musso, Roberto Gambino, and Maurizio Cassader p p p p p p p p p p p p p p p p p p p p p p p p p p p p 361
The Brain-Gut Axis in Abdominal Pain Syndromes
Emeran A. Mayer and Kirsten Tillisch p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 381
Cognitive Therapy: Current Status and Future Directions
Aaron T. Beck and David J.A. Dozois p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 397
Annu. Rev. Med. 2011.62:11-24. Downloaded from www.annualreviews.org
by University of Oxford - Bodleian Library on 05/27/11. For personal use only.
Toward Fulfilling the Promise of Molecular Medicine
in Fragile X Syndrome
Dilja D. Krueger and Mark F. Bear p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 411
Stress- and Allostasis-Induced Brain Plasticity
Bruce S. McEwen and Peter J. Gianaros p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 431
Update on Sleep and Its Disorders
Allan I. Pack and Grace W. Pien p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 447
A Brain-Based Endophenotype for Major Depressive Disorder
Bradley S. Peterson and Myrna M. Weissman p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 461
Indexes
Cumulative Index of Contributing Authors, Volumes 58–62 p p p p p p p p p p p p p p p p p p p p p p p p p p p 475
Cumulative Index of Chapter Titles, Volumes 58–62 p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 479
Errata
An online log of corrections to Annual Review of Medicine articles may be found at
http://med.annualreviews.org/errata.shtml
Contents
vii