phenotypes, such as cachexia and low exercise capacity, have
not been widely analyzed in COPD genetic studies.
Perhaps the greatest problem in the candidate gene era
of COPD genetic studies was improper candidate gene selection, which reflects our limited understanding of COPD
pathogenesis. However, the application of genome-wide association studies (GWAS), which provide an unbiased and
comprehensive search throughout the genome for common
susceptibility loci, has changed the landscape of COPD genetics. Based on GWAS, three genetic loci have been unequivocally associated with COPD susceptibility, located on
chromosome 4 near the HHIP gene, on chromosome 4 in the
FAM13A gene, and on chromosome 15 in a block of genes
which contains several components of the nicotinic acetylcholine receptor as well as the IREB2 gene.
In 2009, a series of studies provided convincing support
for these three genetic loci in COPD susceptibility. Pillai and
colleagues found genome-wide significant associations of the
CHRNA3/CHRNA5/IREB2 region to COPD (12). DeMeo
and colleagues performed gene expression studies of normal
vs. COPD lung tissues followed by genetic association analysis of COPD (13), suggesting that at least one of the key
COPD genetic determinants in the chromosome 15 GWAS
region was IREB2.
In the Framingham Heart Study (14), the HHIP region
was associated with FEV1/FVC at genome-wide significance
with replication of the effect on FEV1/FVC demonstrated in
an independent sample drawn from the Family Heart Study,
and this same region nearly reached genome-wide significance with COPD susceptibility in the Pillai paper (12). Recently, two papers published in Nature Genetics from large
general population samples have provided strong support for
the association of HHIP SNPs with FEV1/FVC (15, 16). One
of these articles, from the CHARGE Consortium, also found
evidence for association of FEV1/FVC with the FAM13A
locus (15), which has been strongly associated with COPD susceptibility (17).
Moreover, several case-control studies from other European populations have replicated these findings by confirming significant associations to the chromosome 15q25
locus (CHRNA3/CHRNA5/IREB2) (18, 19), chromosome
4q31 locus (HHIP) (20, 21), and chromosome 4q22 locus
(FAM13A) (22). Thus, the frustration of inconsistent genetic
association results in COPD from the beginning of the last
decade has been replaced by optimism regarding the likely
importance of the IREB2/CHRNA3/CHRNA5, HHIP, and
FAM13A loci in COPD susceptibility.
Advantages of creating large networks for genetic analysis
There are likely multiple additional COPD susceptibility genetic determinants that have not yet been identified. In
many other complex diseases, the creation of large collaborative consortia has enabled highly powered genome-wide association studies that have led to the identification
of multiple novel genetic susceptibility loci. For example,
a Type 2 Diabetes mellitus consortium performed GWAS
in 8, 130 Cases and 38,987 Controls and identified multiple novel susceptibility loci (23). The International Lung
International COPD Genetics Conference
Cancer Consortium found new SNPs that were associated
with disease in Asian populations (24, 25). The ENGAGE
consortium discovered sequence variants associated with
smoking behavior within regions harboring nAChR genes
(CHRNB3–CHRNA6, 8p11) and a nicotine-metabolizing
enzyme (26). We anticipate that a similar collaborative consortium approach in COPD could lead to the identification
of additional novel COPD genetic determinants.
Gaps in current genetic knowledge
The most fundamental gap in current COPD genetics knowledge is that there are probably many genetic determinants of
COPD, but only three genomic regions likely to contain such
susceptibility loci have been conclusively identified. Moreover, the functional genetic variants within the three existing
COPD GWAS regions remain to be found. To adequately analyze the various subtypes of COPD, studies that include multiple ethnic groups as well as multiple environmental factors
that influence inflammation will be required in large sample sizes. More recently, some studies have combined results
from several populations to increase the numbers of cases
and controls. In more than 8300 subjects in seven study populations, the minor allele of a SNP in MMP12 was associated
with a positive effect on lung function and a reduced risk of
COPD (27). The genome-wide association study that identified FAM13A included three sets of COPD cases and smoking controls (17). However, these studies are still underpowered to identify genetic determinants of small effect, and establishing a consortium of groups studying cigarette smokers
may facilitate pooling large samples to identify genetic variants associated with COPD susceptibility.
GENETIC TECHNOLOGIES AVAILABLE FOR AN
INTERNATIONAL COPD GENETICS CONSORTIUM
It is desirable that the full power of modern genetic and
genomic technology and techniques be brought to bear
on COPD. Statistical genetic approaches should begin with
meta-analyses of currently completed GWA studies, including imputation of polymorphisms from the 1000 Genomes
Project. Analyses should routinely include epidemiologically
important covariates such as sex, age at onset, and smoking history. Ancestry needs to be matched carefully between
cases and controls, using, for example, principal component
analyses. Multi-marker techniques to identify polygenic effects below the GWAS threshold may be useful in identifying
genes and pathways impacting on the disease.
Genome-wide SNP genotyping of several thousand or
more cases is necessary, particularly using existing European
panels of subjects that have not yet been genotyped and cases
and controls of non-European ancestry. It is noted that there
exists a wide range of previously genotyped European controls that could be used wherever possible.
Further meta-analysis of the full dataset should be completed after the additional genotyping. Ideally these results would be integrated with large-scale studies of other
smoking-related diseases (particularly lung cancer and cardiovascular disease), with studies of smoking behavior and
