ARG1 is a novel bronchodilator response gene: screening and replication in four asthma cohorts.
ABSTRACT Inhaled beta-agonists are one of the most widely used classes of drugs for the treatment of asthma. However, a substantial proportion of patients with asthma do not have a favorable response to these drugs, and identifying genetic determinants of drug response may aid in tailoring treatment for individual patients.
To screen variants in candidate genes in the steroid and beta-adrenergic pathways for association with response to inhaled beta-agonists.
We genotyped 844 single nucleotide polymorphisms (SNPs) in 111 candidate genes in 209 children and their parents participating in the Childhood Asthma Management Program. We screened the association of these SNPs with acute response to inhaled beta-agonists (bronchodilator response [BDR]) using a novel algorithm implemented in a family-based association test that ranked SNPs in order of statistical power. Genes that had SNPs with median power in the highest quartile were then taken for replication analyses in three other asthma cohorts.
We identified 17 genes from the screening algorithm and genotyped 99 SNPs from these genes in a second population of patients with asthma. We then genotyped 63 SNPs from four genes with significant associations with BDR, for replication in a third and fourth population of patients with asthma. Evidence for association from the four asthma cohorts was combined, and SNPs from ARG1 were significantly associated with BDR. SNP rs2781659 survived Bonferroni correction for multiple testing (combined P value = 0.00048, adjusted P value = 0.047).
These findings identify ARG1 as a novel gene for acute BDR in both children and adults with asthma.
- [Show abstract] [Hide abstract]
ABSTRACT: Beta2 (β2) adrenergic receptor agonists (beta agonists) are a commonly prescribed treatment for asthma despite the small increase in risk for life-threatening adverse responses associated with long-acting beta agonist (LABA). The concern for life-threatening adverse effects associated with LABA and the inter-individual variability of therapeutic responsiveness to LABA-containing combination therapies provide the rationale for pharmacogenetic studies of beta agonists. These studies primarily evaluated genes within the β2-adrenergic receptor and related pathways; however, recent genome-wide studies have identified novel loci for beta agonist response. Recent studies have identified a role for rare genetic variants in determining beta agonist response and, potentially, the risk for rare, adverse responses to LABA. Before genomics research can be applied to the development of genetic profiles for personalized medicine, it will be necessary to continue adapting to the analysis of an increasing volume of genetic data in larger cohorts with a combination of analytical methods and in vitro studies.Clinical Genetics 03/2014; · 3.65 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: Single nucleotide polymorphisms (SNPs) influence a patient's response to inhaled corticosteroids and β2-agonists, and the effect of treatment with inhaled corticosteroids is synergistic with the effect of β2-agonists. We hypothesized that use of inhaled corticosteroids could influence the effect of SNPs associated with a bronchodilator response. To assess whether, among subjects with asthma, the association of SNPs with bronchodilator response is different between those treated with inhaled corticosteroids versus those on placebo. A genome-wide association analysis was conducted by using 581 white subjects from the Childhood Asthma Management Program. By using data for 449,540 SNPs, we conducted a gene by environment analysis in PLINK with inhaled corticosteroid treatment as the environmental exposure and bronchodilator response as the outcome measure. We attempted to replicate the top 12 SNPs in the Leukotriene Modifier or Corticosteroid or Corticosteroid-Salmeterol Trial. The combined P value for the Childhood Asthma Management Program and Leukotriene Modifier or Corticosteroid or Corticosteroid-Salmeterol Trial populations was 4.8 × 10(-8) for rs3752120, which is located in the zinc finger protein gene ZNF432 and has an unknown function. Inhaled corticosteroids appear to modulate the association of bronchodilator response with variant(s) in the ZNF432 gene among adults and children with asthma.The Journal of allergy and clinical immunology 11/2013; · 12.05 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: Pharmacogenetics is being used to develop personalized therapies specific to subjects from different ethnic or racial groups. To date, pharmacogenetic studies have been primarily performed in trial cohorts consisting of non-Hispanic white subjects of European descent. A "bottleneck" or collapse of genetic diversity associated with the first human colonization of Europe during the Upper Paleolithic period, followed by the recent mixing of African, European, and Native American ancestries, has resulted in different ethnic groups with varying degrees of genetic diversity. Differences in genetic ancestry might introduce genetic variation, which has the potential to alter the therapeutic efficacy of commonly used asthma therapies, such as β2-adrenergic receptor agonists (β-agonists). Pharmacogenetic studies of admixed ethnic groups have been limited to small candidate gene association studies, of which the best example is the gene coding for the receptor target of β-agonist therapy, the β2-adrenergic receptor (ADRB2). Large consortium-based sequencing studies are using next-generation whole-genome sequencing to provide a diverse genome map of different admixed populations, which can be used for future pharmacogenetic studies. These studies will include candidate gene studies, genome-wide association studies, and whole-genome admixture-based approaches that account for ancestral genetic structure, complex haplotypes, gene-gene interactions, and rare variants to detect and replicate novel pharmacogenetic loci.The Journal of allergy and clinical immunology 01/2014; 133(1):16-26. · 12.05 Impact Factor
ARG1 is a novel bronchodilator response gene: screening and replication in
four asthma cohorts
Augusto A. Litonjua1,2,3, Jessica Lasky-Su1,3,4, Kady Schneiter5, Kelan G. Tantisira1,2,3, Ross
Lazarus1,3, Barbara Klanderman1,3, John J. Lima6, Charles G. Irvin7, Stephen P. Peters8, John P.
Hanrahan9, Stephen B. Liggett10, Gregory A. Hawkins8, Deborah A. Meyers8, Eugene R.
Bleecker8, Christoph Lange4, Scott T. Weiss1,3
1. Channing Laboratory, Brigham and Women’s Hospital and Harvard Medical School, Boston,
2. Pulmonary Division, Brigham and Women’s Hospital and Harvard Medical School, Boston,
3. Center for Genomic Medicine, Brigham and Women’s Hospital, Boston, MA
4. Harvard School of Public Health, Boston, MA
5. Department of Mathematics and Statistics, Utah State University, Logan, Utah
6. Nemours Children’s Clinic, Centers
Pharmacogenetics, Jacksonville, FL
7. Vermont Lung Center, Department of Medicine and Physiology, University of Vermont,
8. Center for Human Genomics, Section of Pulmonary, Critical Care, Allergy and Immunologic
Diseases, Wake Forest University School of Medicine, Winston Salem, NC
9. Pulmonary Clinical Research, Sepracor Inc., Marlborough, MA
10. Cardiopulmonary Genomics Program, University of Maryland School of Medicine,
for Clinical Pediatric Pharmacology &
Correspondence: Augusto A. Litonjua, M.D., M.P.H.
181 Longwood Avenue
Boston, MA 02115
Acknowledgement and Support: This work was supported by U01 HL65899: The
Pharmacogenetics of Asthma Treatment from the NHLBI. We thank all families for their
enthusiastic participation in the Camp Genetics Ancillary Study, supported by the National
Heart, Lung, and Blood Institute, NO1-HR-16049. We acknowledge the CAMP investigators
and research team, supported by NHLBI, for collection of CAMP Genetic Ancillary Study data.
Additional support for this research came from grants N01 HR16044, HR16045, HR16046,
HR16047, HR16048, HR16049, HR16050, HR16051, and HR16052 from the National Heart,
Lung and Blood Institute. All work on data from the CAMP Genetics Ancillary Study was
conducted at the Channing Laboratory of the Brigham and Women’s Hospital under appropriate
CAMP policies and human subjects protections. We acknowledge the American Lung
Association (ALA) and the ALA’s Asthma Clinical Research Centers investigators and research
teams for use of LOCCS and LoDo data, with additional funding from HL071394 and
HL074755 from the NHLBI, and Nemours Children's’ Clinic. GlaxoSmithKline supported the
AJRCCM Articles in Press. Published on July 10, 2008 as doi:10.1164/rccm.200709-1363OC
Copyright (C) 2008 by the American Thoracic Society.
conduct of the LOCCS Trial by an unrestricted grant to the ALA. We acknowledge Sepracor,
Inc. for use of the Asthma Trial data.
Short Running Head: ARG1 and bronchodilator response in asthma
Subject Category: 58 (Asthma genetics) and 168 (Genetic epidemiology)
Text Word Count: 3,307
AT A GLANCE COMMENTARY
Scientific Knowledge on the Subject
Investigations on asthma pharmacogenetics to date have mostly studied only one or a few SNPs
in the ß2-adrenergic receptor gene (ADRB2). Since response to inhaled ß agonists in asthma is a
complex phenotype, it is likely that other genes are involved.
What This Study Adds to the Field
This study identifies the arginase1 gene (ARG1) as a potential ß agonist response gene, using a
novel family-based method to screen variants in genes in the steroid and ß adrenergic pathways.
Findings were replicated in 3 separate asthma populations.
This article has an online data supplement, which is accessible from this issue's table of content
online at www.atsjournals.org
Rationale: Inhaled ß agonists are one of the most widely used classes of drugs for the treatment
of asthma. However, a substantial proportion of asthmatics do not have a favorable response to
these drugs, and identifying genetic determinants of drug response may aid in tailoring treatment
for individual patients. Objective: To screen variants in candidate genes in the steroid and ß
adrenergic pathways for association with response to inhaled ß agonists. Methods: We
genotyped 844 single nucleotide polymorphisms (SNPs) in 111 candidate genes in 209 children
and their parents participating in the Childhood Asthma Management Program. We screened the
association of these SNPs with acute response to inhaled ß agonists (bronchodilator response,
BDR) using a novel algorithm implemented in a family-based association test that ranked SNPs
in order of statistical power. Genes that had SNPs with median power in the highest quartile
were then taken for replication analyses in three other asthma cohorts. Results: We identified 17
genes from the screening algorithm and genotyped 99 SNPs from these genes in a second
population of asthmatics. We then genotyped 63 SNPs from 4 genes with significant associations
with BDR, for replication in a third and fourth population of asthmatics. Evidence for association
from the 4 asthma cohorts was combined, and SNPs from ARG1 were significantly associated
with BDR. SNP rs2781659 survived Bonferroni correction for multiple testing (combined p-
value = 0.00048, adjusted p-value = 0.047). Conclusion: These findings identify ARG1 as a
novel gene for acute BDR in both childhood and adult asthmatics.
Word count: 251
Key Words: Pharmacogenetics; Asthma; Bronchodilator Agents
Asthma is a complex genetic disorder that currently affects about 300 million people
worldwide(1). Asthma remains the most common chronic disease of childhood in the developed
world(2, 3), and incurs a significant healthcare cost(4). β-agonists form one of the oldest classes
of drugs in medicine(5). They are the most effective medications for the treatment of acute
asthma and remain one of the cornerstones of chronic asthma therapy. However, variability in
the response to inhaled β-agonists exists(6), and it has been estimated that a substantial
proportion of that response is genetic in nature.
To date, asthma pharmacogenetic studies in general(7), and response to inhaled
bronchodilators in particular(8), have been based on one or more polymorphisms in a single
gene. Recent studies from the Asthma Clinical Research Network, for instance, have reported
adverse effects of regular albuterol treatment among asthmatics who were homozygous for the
+49 A allele (Arg16) of the ADRB2 gene(9, 10). However, because asthma is a complex disorder
and response to inhaled beta-agonist drugs is a complex phenotype, it is likely that other genes
also impact on this phenotype. When more than one or a few polymorphisms are tested, the
chances of obtaining false positives increases – the multiple testing issue – and methods to
adjust for the total number of tests need to be applied. Additionally, the ideal design for these
studies would employ samples with large numbers (i.e. in the thousands) of subjects that would
help to offset the more stringent statistical criteria. Unfortunately, most existing datasets
available for asthma pharmacogenetics are of modest size, and thus, screening methods to limit
the number of tests in the first stages of the analysis can help alleviate the multiple testing
We conducted an analysis to screen 844 single nucleotide polymorphisms (SNPs) from
111 candidate genes for association with bronchodilator response (BDR) to inhaled β-agonist in
an asthma clinical trial cohort. Because the issue of multiple testing , we employed an algorithm
in a family-based association testing framework that allowed us to screen SNPs based on power
for replication(11). This screening methodology allows for the identification of the most
promising SNPs for testing without biasing the nominal significance level of the test statistic,
and recently has led to the identification of disease-susceptibility genes(12-14). Additionally, this
algorithm allowed us to screen and test in the same population. After identifying the most
promising SNPs, we then attempted replication in three additional asthma clinical trial cohorts.
Some of the results from these analyses have been previously reported in abstract form(15)
We utilized DNA samples from four clinical trials. All patients or their legal guardians
consented to each trial study protocol and ancillary genetic testing. The population we used for
the screening algorithm was the Childhood Asthma Management Program (CAMP). Trial
design and methodology have been published(16, 17). A total of 209 Caucasian probands
(randomized to the placebo group) and their parents were included as part of parent-child trios
for the screening analyses. Only subjects randomized to the placebo group were utilized for the
screening analyses to avoid confounding effects of medications (corticosteroids and nedocromil)
other than inhaled β-agonist.
The population we used for the first replication study (hereafter called the Asthma Trial)
was composed of 432 Caucasian subjects with asthma(18, 19) who were part of an asthma
medication trial conducted by Sepracor, Inc. in the United States. Two completed trials
conducted by the American Lung Association Asthma Clinical Research Centers (ALA-ACRC),
the Leukotriene modifier or Corticosteroid or Corticosteroid Salmeterol trial (LOCCS)(20) and
the Effectiveness of Low Dose Theophylline as Add-on Treatment in Asthma (LODO) trial(21),
were used as the second and third replication samples. The 166 Caucasian subjects from the
LOCCS trial and the 155 Caucasian subjects from the LODO trial for whom DNA was available
were used for this analysis. Detailed information on these subjects has been previously published
and is included in the online data supplement.
Selection of Genes and SNPs; Genotyping.
We genotyped 844 SNPs in 111 candidate genes: 42 genes involved in beta adrenergic
signaling and regulation; 28 genes involved in innate glucocorticoid synthesis and metabolism,
cellular receptors, and transcriptional regulators; and 41 genes from prior asthma association
studies that had been previously conducted in the CAMP dataset (Table E1 in the online data
supplement). Candidate genes in the beta adrenergic and corticosteroid pathways were selected
based on prior studies in the literature, their known involvement in metabolic pathways(22, 23),
and on expert opinion (SBL and KGT). Corticosteroid pathway candidate genes were included in
this analysis because of the known interactions between beta2 agonists and corticosteroids(24,
25). Finally, we included the candidate genes that our group had previously genotyped and
studied in CAMP, since these were already available and so as to appropriately adjust our current
analyses for all prior tests conducted with these genes. SNPs were primarily selected utilizing
public databases, although resequencing of several core genes was performed. We over-sampled
exonic and promoter regions and attempted coverage of at least one SNP every 10 kb. We
emphasized golden-gate validated and LD tag SNPs, where available.
SNPs were genotyped via an Illumina BeadStation 500G (Illumima Inc., San Diego, CA)
and via a SEQUENOM MassARRAY MALDI-TOF mass spectrometer (Sequenom, San Diego,
CA). Further details are included in the online data supplement. SNPs were also checked for
Mendelian inconsistencies and for Hardy-Weinberg Equilibrium.
The primary outcome measure of the association analyses was acute response to inhaled
bronchodilator (BDR), and calculated as the percent difference between the pre- and post-
bronchodilator FEV1 value (BDR = 100 x [post FEV1- pre FEV1/pre FEV1]). In all analyses,
both screening and replication, BDR was treated as a continuous variable. We initially screened
the genotypic association with BDR in CAMP using a modified version of the screening
algorithm as detailed by Van Steen, et al(11). Further details are included in the online data
supplement. The rationale for using CAMP data as the screening set is because the screening
methodology was designed for family data. No screening methodology has yet been published
for population-based data. We used the 11 repeated measures of BDR over the four years of the
trial in the Placebo group using the FBAT-PC statistic(26) to maximize the heritability of a given
marker and thereby maximize power for the screening stage. Only additive genetic models were
evaluated and all analyses in the screening stage were adjusted for age, sex, height, and baseline
FEV1. We selected the most powerful candidate genes by first ranking the individual SNPs based
upon conditional power and then evaluating the median rank of all of the SNPs within a given
candidate gene. We selected genes whose median SNP ranks for power were within the top 25%
of all SNPs genotyped to be taken forward for genotyping in the Asthma Trial population. For
those selected SNPs, we evaluated the FBAT-PC statistics for directionality of the association
and p-value for each SNP, and this allowed us to conduct 1-sided tests in the replication
analyses. We selected 99 SNPs from 17 genes for replication in the Asthma Trial population.
The analyses of the replicate populations were performed using generalized linear models
as incorporated into PROC GLM of the SAS statistical analysis software (version 9.0, SAS
Institute, Cary, NC), and SNP genotypes were coded for additive models. Only BDR calculated
from spirometric measurements at all baseline visits (prior to randomization) for all the trials
were used. All analyses in the replication populations adjusted for age, sex, height, and baseline
FEV1. Genes with at least one SNP that was at least marginally associated with BDR (one-sided
p-value < 0.05) were then genotyped in the final two replicate populations. For each SNP where
the direction of the association was in the same direction in each of the four populations, we then
combined the p-values (2-sided) from the original family-based analysis and the 1-sided p-values
from the replication cohorts using Fisher’s method(27) to increase statistical efficiency(28). No
evidence for population stratification was found in any of the three populations. Further details
on analytic issues are included in the online data supplement.
The baseline characteristics of participants of the 4 asthma cohorts are shown in Table 1.
The CAMP subjects on whom we performed the initial screen were composed of children,
whereas the three replication cohorts were primarily adult asthmatics. In the CAMP subjects, we
screened 844 SNPs from 111 candidate genes (Supplementary Table 1 and Fig. 1). In the
screening analysis, we ranked the individual SNPs based on the highest to lowest power
estimates from the FBAT screening analysis. From these, we identified 19 genes whose median
SNP ranks for power were within the top 25% of all SNPs genotyped (Supplementary Table 1).
Because the family-based analysis suggested directionality of the association, we conducted 1-
sided tests in the replication datasets. The first replication analysis was conducted on 432
Caucasian adult asthmatics who had participated in a clinical trial of an asthma medication
(Asthma Trial). We successfully genotyped 99 SNPs (11.7% of all the SNPs that were screened)
from 17 genes in the Asthma Trial. In this first replication analysis, 9 genes contained at least 1
SNP that was at least marginally (1-sided p ≤ 0.05) associated with BDR. We then genotyped 63
SNPs from these 9 genes and tested them in the final two asthma clinical trial populations from
the American Lung Association Asthma Clinical Research Network: the LOCCS trial and the
Table 2 summarizes the results of the replication analyses. Several SNPs were
individually associated with BDR in the four populations. The p-values from each of these
populations were combined, and four SNPs from ARG1 (rs2781659, rs2781663, rs2781665, and
rs2749935) showed the strongest evidence for association with BDR. After applying Bonferroni
correction for the 99 tests in the initial replication analysis, SNP rs22781659 remained
significantly associated with BDR (combined p-value=0.00048; Bonferroni-corrected p-
value=0.047). Evidence for association of SNPs rs2781663 and rs2781665 was borderline
significant after adjustment for multiple testing (Bonferroni-corrected p-values 0.075 and 0.085,
We examined the effect of each of these 3 SNPs on the magnitude of BDR in each of the
populations (Table 3). In each case, the presence of the minor allele was associated with lower
adjusted BDR compared with the homozygous major allele, consistent with the initial FBAT-PC
results. Figure 2 compares the LD patterns of the 5 ARG1 SNPs in the 4 asthma populations by
plotting pairwise r2 in physical order. The pairwise r2 values are similar in each of the
populations. The 3 SNPs that were associated with BDR (rs2781659, rs2781663, and rs2781665)
were in strong LD with each other (r2 values ranging from 95% to 100%, depending on the
population). For example, in the CAMP population, r2 between SNPs rs2781659 and rs2781663
was 100%, while the r2 between SNPs rs2781659 and rs2781665 was 99%. In contrast, the other
SNP that was more weakly associated with BDR, rs2749935, was not as tightly linked with the
other 3 SNPs (r2 was 55% with rs2781659, rs2781663, and rs2781665).
Prior investigations into the pharmacogenetics of asthma have generally been limited to
one or a few SNPs from one gene. We investigated 844 SNPs from 111 candidate genes selected
from asthma β-agonist and corticosteroid pathways, and from our prior candidate gene studies,
and screened these SNPs for association with BDR using a family-based screening algorithm
which allowed us to rank the SNPs based on estimated power for replication. We then genotyped
99 SNPs from 17 genes in a population-based cohort of asthmatics, who participated in an
asthma clinical trial. Finally, we genotyped 83 SNPs in 7 genes in two separate cohorts of
asthmatics. We found SNPs in the ARG1 gene to be associated with BDR in these three asthma
populations, after adjusting for multiple comparisons.
ARG1 has recently been implicated in asthma. Zimmerman et al(29) reported increased
expression of ARG1 and ARG2 in murine lung, and also found increased arginase 1 protein
expression from human asthma bronchoalveolar lavage cells. Variants in ARG1 were associated
with atopy in a cohort of Mexican asthmatics(30). ARG1 maps to chromosome 6q23 and encodes
one isoform of the enzyme arginase, which metabolizes L-arginine. L-Arginine homeostasis is
involved in the regulation of airway function, since the availability of this amino acid to nitric
oxide synthase (NOS) determines the production of the endogenous bronchodilator nitric oxide
(NO)(31). Changes in L-arginine homeostasis may contribute to many of the features of asthma,
such as airway hyperresponsiveness, airway inflammation, and airway remodelling(32).
Intracellular L-arginine levels are regulated by at least 3 distinct mechanisms (reviewed by
Maarsingh et al(32)): (i) cellular uptake by cationic amino acid transporters, (ii) recycling from
L-citrulline, and (iii) metabolism by NOS and arginase. Arginase is postulated to be involved in
asthma by depleting stores of L-arginine, a NOS substrate, which leads to decreased production
of NO, a potent bronchial smooth muscle relaxer(33, 34), and it has been shown to inhibit airway
smooth muscle relaxation(35, 36). Finally, RNA interference of arginase1 in the lungs resulted in
complete loss of airway hyperresponsiveness to methacholine due to IL-13 treatment(37). This
correlated with arginase 1 expression, which suggests that the polymorphismsms involved with
the current findings in human asthma may cause a loss of expression or function of arginase 1.
We used a gene-based strategy to select SNPs to take forward for replication. In this
method, after ranking SNPs from 1 (most power) to 844 (least power), we grouped all SNPs for
each gene and calculated the median SNP rank for that gene. Thus, while some genes had one or
two SNPs that were assigned high ranks, these genes may not be taken forward because the
median SNP rank did not meet the predetermined cutoff. We adopted this strategy since we were
not sure that LD patterns across the 4 asthma populations would be similar. It is interesting to
note that ADRB2, a gene that has been widely studied in asthma pharmacogenetics(38), was not
one of the genes that was selected using this strategy, despite including 18 SNPs from this gene
in the screening analysis. It is possible that there was insufficient power in the screening stage
since we only analyzed the 209 trios in the placebo group in CAMP. However, it should be noted
that a prior analysis using all 400 trios also did not find an association with any of the ADRB2
SNPs and BDR(39). Furthermore, the phenotype that we investigated is different from that
reported in other studies reporting on the pharmacogenetic effect of ADRB2(9, 10). We are
currently performing additional genotyping and analyses using a SNP-based strategy for
replication, rather than the gene-based strategy that we used here, to see if we identify important
SNPs in this gene and others for association with BDR.
Our analysis used the phenotype of acute response to a short-acting β2-agonist, albuterol,
in part because this was the phenotype that was common to all asthma cohorts. In the screening
algorithm, we used the information from repeated measures of BDR among the 209 Caucasian
children randomized to the placebo group in the CAMP study over the four years of the trial.
This was done to increase the power for the screening method. In contrast, for the replication
cohorts, we only used the information on BDR response on entry into the respective studies, in
order to standardize the phenotype. Thus, our results may not be applicable to asthma patients
who are on regular β2-agonist treatment (either short- or long-acting). We also did not address
interactions with any other class of asthma medication, since baseline medication was different
for all the populations: BDR was performed in both CAMP and the Adult Trial populations after
several weeks of being off all asthma medications; LOCCS subjects were on inhaled
corticosteroids for 4-6 weeks prior to BDR testing; and drug regimen for LODO subjects were
not changed prior to entry into the trial.
We employed a novel method of screening a large number of SNPs for association
analysis(11). This method has been successfully used to identify disease-susceptibility genes(12-
14). Since this method has only been developed for family-based studies and not population-
based studies, we used the CAMP population for screening the original 844 SNPs. The
traditional method would have been to analyze all the SNPs in one population, determine which
SNPs were associated with BDR at a predetermined level of significance, then test these SNPs in
the replication populations. However, if we had employed this usual method for gene finding, we
would then have to adjust our overall results for the 844 SNPs that were originally tested, and it
is likely that no finding would have survived this adjustment for multiple testing, even if the
association was real. In our method, because we screened on power and not p-value, we only
needed to adjust for the 99 tests in the first replication step. Thus, this screening method allows
the use of modest sized populations for gene discovery because it limits the number of tests that
are actually being performed.
The population to which we applied our screening algorithm was a cohort of childhood
asthmatics, whereas the three asthma replication cohorts were composed predominantly of adult
asthmatics. As we stated previously, the rationale for this is that the screening method was
developed for the setting of family-based studies and not for population-based studies. There is
no similar screening method that has yet been developed for population-based studies. Because
our replication populations were of small to modest sizes, we applied the screening method as a
means of minimizing the number of tests. While there were only 209 parent-child trios included
in the screening analysis, we maximized the power in the screening stage by utilizing the 11
repeated measures of BDR over the 4 years of the trial. Additionally, there were differences in
the asthma severity and in the magnitude of the BDR between the populations as shown in Table
1. Despite these differences, we were able to detect associations between SNPs in ARG1 and
BDR in each of the three replication populations. While the association between these SNPs and
BDR in CAMP were not statistically significant, the effect sized of each SNP were of sufficient
magnitude for them to be selected based on power in the screening analysis. We can only
surmise at this point, that there may be age-related effects associated with the SNPs in this gene.
We believe, therefore, that the results of the association between ARG1 polymorphisms and BDR
are robust and applicable to both childhood and adult asthmatics in a variety of settings.