Pharmacogenetics of Asthma

Lyle J. Palmer, Eric S. Silverman, Scott T. Weiss, and Jeffrey M. Drazen

Channing Laboratory, Department of Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts; Department of Epidemiology and Biostatistics, Case Western Reserve University, Cleveland, Ohio; and Pulmonary Division, Department of Medicine, Brigham and Women's Hospital, Boston, Massachusetts

	INTRODUCTION

TOP INTRODUCTION PHARMACOGENETIC PATHWAYS AND... GENETIC ASSOCIATION ANALYSIS... PREVIOUS STUDIES OF ASTHMA... STATISTICAL ISSUES FUTURE DIRECTIONS AND ISSUES CONCLUSIONS REFERENCES

Asthma is a common chronic disease of unknown cause, for which a number of pharmacological treatments have been developed. These treatments have a modest efficacy overall, due in part to widely variable individual responses to asthma drugs. Because of such variability, it is clear that some of the substantial resources expended on asthma medication, estimated to exceed US$3 billion per annum in the United States alone (1), would be better spent targeting those patients who would benefit the most. At present there are no proven methods of effectively predicting response and prospectively targeting asthma treatment.

Asthma is a syndrome rather than a distinct disease, and probably has multiple environmental and genetic determinants (2). In genetic parlance, this is known as a "complex" phenotype. A component of this complexity is a highly variable response to pharmacological therapy among individual patients with asthma (3, 4). Pharmacogenetics is the study of the role of genetic determinants in the variable response to therapy. Ideally, we would be able to stratify a population needing treatment into those likely, or unlikely, to respond to treatment as well as those likely, or unlikely, to experience adverse side effects. This Perspective summarizes the State of the Art in asthma pharmacogenetics and suggests future directions and contributions of pharmacogenetics to our understanding and treatment of asthma.

	PHARMACOGENETIC PATHWAYS AND PHENOTYPES

TOP INTRODUCTION PHARMACOGENETIC PATHWAYS AND... GENETIC ASSOCIATION ANALYSIS... PREVIOUS STUDIES OF ASTHMA... STATISTICAL ISSUES FUTURE DIRECTIONS AND ISSUES CONCLUSIONS REFERENCES

There are four major classes of asthma pharmacotherapy currently in widespread use (5): (i) ₂-agonists (-agonist) used by inhalation for the relief of airway obstruction (e.g., albuterol, salmeterol, fenoterol); (ii) glucocorticosteroids for both inhaled and systemic use (e.g., beclomethasone, triamcinolone, prednisone); (iii) theophylline and its derivatives, used for both the relief of bronchospasm and the control of inflammation; and (iv) inhibitors and receptor antagonists of the cysteinyl-leukotriene pathway (e.g., montelukast, pranlukast, zafirlukast, zileuton). Although there are data on the variability of the treatment response for each of these classes of agents, there are no systematic studies of the reasons for variance in the treatment response to steroids or theophylline. Therefore, this perspective focuses on the specific pharmacogenetics of -agonists and inhibitors of the cysteinyl pathway (4) and on general considerations related to pharmacogenetic mechanisms.

Variability in individual asthma treatment response may be due to many factors, including the severity and type of disease, treatment compliance, intercurrent illness, other medication taken (drug-drug interaction), environmental exposures, and age (3). However, there is reason to believe that genetic factors underlie much of the observed treatment variance. Comparison of the inter- and intraperson variance in treatment response has suggested that up to 80% of such variance in white individuals may have a genetic basis (4). Although many pharmacogenetic mechanisms are possible, genetic variants may alter response to drugs in three main ways (Table 1):

1. Variation in metabolism of a drug among individuals, especially in enzymes involved in the catabolism or excretion of a drug: an important example is the highly genetically diverse cytochrome P-450 system, known to have many pharmacogenetic effects. No genes have yet been identified as responsible for this kind of pharmacogenetic mechanism in asthma. Although the variations observed in theophylline response among patients with asthma may result from variance in the catabolism of theophylline, there have been no genetically based population studies of such a mechanism, and the genes involved remain unknown.

2. Variance among population members with respect to drug adverse effects that are not based on the drug's action: the most striking pulmonary example is the variation in the metabolism of isoniazid and its side effects (6), but no studies of the side effects of asthma treatments have established a genetic basis. It is interesting to speculate that the variation in incidence of adverse effects of inhaled glucocorticoids (e.g., glaucoma, cataracts, or the rate of bone loss) may be genetically determined, but there have been no data establishing a specific gene or locus associated with these adverse effects.

3. Genetic variance in the drug treatment target or target pathways: all of the currently available data on asthma pharmacogenetics fall into this single mechanistic category, in which a population is conceptually divided into responders and nonresponders, and analysis of specific DNA variants is used in an attempt to distinguish these groups.

Before proceeding further, we briefly define and explain genetic methods and terminology used in this field.

	GENETIC ASSOCIATION ANALYSIS USING SINGLE-NUCLEOTIDE POLYMORPHISMS

TOP INTRODUCTION PHARMACOGENETIC PATHWAYS AND... GENETIC ASSOCIATION ANALYSIS... PREVIOUS STUDIES OF ASTHMA... STATISTICAL ISSUES FUTURE DIRECTIONS AND ISSUES CONCLUSIONS REFERENCES

Genetic polymorphism arises from mutation. The simplest class of polymorphism derives from a single-base mutation that substitutes one nucleotide for another, termed a "single-nucleotide polymorphism", or SNP (pronounced "snip") (7). SNPs are recognized through a variety of techniques that exploit the known DNA sequence variant (7). SNPs may be found in coding or regulatory regions of a gene and thus can directly affect gene function or expression. However, most SNPs do not alter gene structure or function in any way, and therefore may not be directly associated with any change in phenotype. Thus, it is important to ascertain whether the DNA sequence variant under consideration is potentially functional (i.e., could lead to the observed biology) or is a marker in linkage disequilibrium with another DNA sequence variant that is the actual cause of the variable treatment response.

Because of their potential biological importance, the common SNPs in the human genome increasingly have been the subject of large-scale cataloguing projects funded by both government and industry groups (8). Alhough these enterprises are constructing large SNP databases (9), this process is not yet complete. Many investigators interested in a specific pathway have independently sought to identify sequence variants.

There are a number of potential advantages in the use of SNPs to investigate the pharmacogenetics of complex human diseases such as asthma (9, 10). First, SNPs are the most common type of polymorphism across the human genome and are found in exons, introns, promotors, enhancers, and intergenic regions. Precise estimates of SNP frequency are difficult to determine and often vary across different populations and genomic regions; however, some studies have suggested that SNPs can be found, on average, every 0.3 to 1 kilobase (kb) within the human genome (11). Second, groups of adjacent SNPs may exhibit patterns of linkage disequilibrium that could be used to enhance gene mapping. Finally, there is good evidence that SNPs are less mutable than other types of polymorphisms (12), allowing more consistent estimates of linkage disequilibrium and gene-phenotype associations. Problems in SNP association analyses arise from technical limitations in genotyping and from the now well-described general limitations of investigating gene-phenotype associations in complex human diseases involving multiple interacting genetic and environmental factors (9, 13).

Although less common than SNPs, another type of genetic polymorphism is the variable nucleotide tandem repeat (VNTR), also termed "microsatellite." This is a group of DNA bases, ranging from a dinucleotide to a heptanucleotide (or larger structure), that is repeated at a particular locus. VNTRs have a large number of alleles (i.e., repeat lengths), and the mutation rate is generally higher than that found in SNPs. If the VNTR is functional, for example, a triplet repeat leading to the insertion or deletion of a repeated amino acid in a protein, it could have a pharmacogenetic effect. Both SNPs and VNTRs that have pharmacogenetic effects in asthma have been identified.

	PREVIOUS STUDIES OF ASTHMA PHARMACOGENETICS

TOP INTRODUCTION PHARMACOGENETIC PATHWAYS AND... GENETIC ASSOCIATION ANALYSIS... PREVIOUS STUDIES OF ASTHMA... STATISTICAL ISSUES FUTURE DIRECTIONS AND ISSUES CONCLUSIONS REFERENCES

Pharmacogenetics of -Agonists in Asthma

There are at least two reasons for the intense interest in the ₂-adrenergic receptor (₂AR) gene and its relationship to treatment response. First, -agonists are the most commonly prescribed asthma medications. Second, there is great controversy among clinicians as to the toxicity and appropriate use of these drugs.

The pharmacodynamics and molecular biology of the possible pathways involved in the action of -agonists have been extensively reviewed (14). -Agonists act via binding to the ₂AR, a cell surface G protein-coupled receptor. Responses to this drug are currently the most investigated pharmacogenetic pathway in asthma (15-17).

Efforts to explain not only individual differences in response to -agonist, but also sporadic reports of tachyphylaxis, have focused on the ₂AR gene, an intronless gene on chromosome 5q31-32, because of its direct interaction with -agonists and its central role in the -agonist pathway (18). A total of 13 polymorphisms in the gene and its transcriptional regulator  upstream peptide have been identified (18). Three closely linked polymorphisms, two coding block SNPs at amino acid positions 16 and 27 and an SNP in the  upstream peptide, are common (i.e., allele frequency > 0.15) in the general white population (19, 20).

The coding variants (at positions 16 and 27) within the ₂AR gene have been shown in vitro to be functionally important (15, 21). The Gly-16 receptor exhibits enhanced downregulation in vitro after agonist exposure (21). In contrast, Arg-16 receptors are more resistant to downregulation. Because of linkage disequilibrium, individuals who are Arg/Arg at position 16 are much more likely to be Glu/Glu at position 27; individuals who are Gly/Gly at position 16 are much more likely to be Gln/Gln at position 27. The position 27 genotypes influence but do not abolish the effect of the position 16 polymorphisms with regard to downregulation of phenotypes in vitro (15, 21). Although initial studies suggested a relationship between the Gly-16 polymorphism and increased risk of severe asthma (22) and increased airway responsiveness (23), subsequent associations with clinical asthma, atopy, and airway responsiveness have been inconsistent (24).

Earlier studies of treatment response phenotypes were either negative (25) or were limited by small numbers of subjects (26). The largest study to date is based on a multicenter, placebo-controlled, double-blind trial run over 16 weeks and involving 255 subjects with mild asthma who were randomized to receive either two puffs of albuterol four times a day on a regularly scheduled basis or treatment only as needed. The initial data for all patients in the trial suggested that, despite the roughly six-puff daily difference in inhaled albuterol between the two treatment groups, there was no difference in A.M. and P.M. peak expiratory flow in the primary trial. These investigators concluded that regular use of albuterol was not more strongly associated with adverse events than use of albuterol as needed (27). However, when the results from 190 of the 255 randomized subjects were stratified by genotype at the position 16 and 27 polymorphisms, a decrease in morning peak expiratory flow was noted among the patients who were Arg/ Arg homozygotes at position 16 and who regularly used albuterol (16, 28). At the end of the four-week runout period, in which all patients used albuterol only as needed, the patients with the Arg/Arg genotypes who had regularly used albuterol during the trial had a morning peak expiratory flow of 30.5 ± 12.1 L/minute lower than that of Arg/Arg patients who had used albuterol on an as-needed basis during the entire trial. The difference between the Arg/Arg regular albuterol users and the Gly/Gly regular albuterol users was roughly 20 L/minute (Figure 1).

View larger version (21K): [in this window] [in a new window]   Figure 1.   2AR position 16 genotype is associated with response to chronic -agonist treatment. Shown is the time course of the change in morning peak expiratory flow (A.M. PEF) among different position 16 genotypes in response to -agonist treatment. Over the treatment and runout period, Arg/Arg patients who received regularly scheduled -agonist treatment (Arg/Arg-Regular) experienced a 30.5 ± 12.1-L/ minute decline in A.M. peak expiratory flow relative to those who received as-needed treatment (Arg/Arg-As needed ) (p = 0.01). Arg16Gly-Gly/Gly patients were not affected by regular treatment (Gly/Gly-Regular). Thus, regular treatment was associated with a 23.8 ± 9.5-L/ minute decline in peak expiratory flow in response to -agonist treatment (Arg/Arg patients) relative to 16 Gly/Gly patients (p = 0.01). Run out = predetermined four-week period when regular -agonist use had been discontinued. Reprinted by permission from Israel and coworkers (16).

Liggett has proposed an explanation that synthesizes the in vitro data and the results from this clinical trial in a so-called dynamic model of receptor kinetics (17). Under this theory, Gly/Gly homozygous individuals are already downregulated as a result of exposure to endogenous cathecholamines. Thus, the tachyphylaxis caused by recurring exogenous exposure to -agonist would be more apparent in the Arg/Arg patients because their receptors had not yet been downregulated. In this model, the initial response to albuterol in -agonist-naive patients would be depressed in those who are Gly/Gly homozygous, because their receptors had been endogenously downregulated to a greater extent than the receptors of patients who are Arg/Arg. The issue of an acute response to -agonist was not addressed in the clinical trial.

Data obtained by Martinez and coworkers (29) on the bronchodilator response after administration of a single dose of albuterol address the issue of acute response to -agonist and support Liggett's idea. The study group consisted of 191 normal children and 78 children with a history of wheezing (37 of whom had a diagnosis of asthma). Both the children with asthma and the normal children, who were -agonist naive, showed a significantly greater percentage of bronchodilator responses in the homozygous Arg-16 group. When the groups were compared, the homozygous Arg-16 children were 5.3-fold more likely to manifest a positive bronchodilator response to albuterol than were Gly-16/Gly-16 children.

Although the dynamic model of receptor kinetics may explain the results of both clinical trials and epidemiologic studies, other explanations are still possible. It may be that other genes acting in concert with the ₂AR are important in determining pharmacologic response in this pathway. Alternatively, although linkage disequilibrium at position 27 cannot explain the results of Israel and coworkers, as both the position 16 and 27 polymorphisms were investigated (16, 28), they did not genotype polymorphisms in the 5' leader cistron, which is known to be in linkage disequilibrium with the position 16 polymorphism and may influence expression of the ₂AR.

The issue of haplotypes remains unresolved. Haplotypes are linear combinations of SNPs along a chromosome; definition of haplotypes among SNPs within a gene may enhance our ability to detect phenotype/genotype correlations. Drysdale and colleagues investigated molecular haplotypes of the 13 SNPs in the promoter and coding regions (30). A study of common haplotypes of these SNPs in 121 white patients with asthma found that certain haplotypes appeared to affect receptor function differentially and also appeared to correlate with clinical phenotypes (30). Although this approach is likely to be more powerful than focusing on a single SNP locus, it does effectively divide the population into multiple small groups, thus requiring large sample sizes to identify a biological effect.

Pharmacogenetics of Leukotrienes in Asthma

The leukotrienes are a family of polyunsaturated lipoxygenated eicosatetraenoic acids that are derived from arachidonic acid and exhibit a wide range of pharmacological and physiological actions (31). Of the three enzymes exclusively involved in the formation of the leukotrienes (5-lipoxygenase [ALOX5], leudotriene C₄ [LTC₄] synthase, and LTA₄ epoxide hydrolase), ALOX5 is the enzyme required for the production of both the cysteinyl-leukotrienes (LTC₄, LTD₄, and LTE₄) and LTB₄. ALOX5 activity in part determines the level of bronchoconstrictor leukotrienes present in the airways, and pharmacological inhibition of the action of ALOX5 or antagonism of the action of the cysteinyl-leukotrienes at their receptor is associated with an amelioration of asthma (32, 33).

The ALOX5 gene promoter contains numerous consensus binding sites for many known transcription factors. Transcriptional activation of the ALOX5 gene has been shown to be dependent in part on transcription factor binding to an Sp-1 binding motif, which is polymorphic and has between three and six tandem repeats in white and black subjects (34). These mutations were found to have significant functional consequences in the context of promoter-reporter constructs (35), such that constructs containing more or fewer copies of the VNTR compared with the most commonly occurring five-repeat constructs were found to have diminished activity. This led to the hypothesis that patients with VNTRs other than the wild type, that is, five repeats of the sequence -GGGCGG- in the core promoter, would have diminished transcription of the ALOX5 gene and hence produce fewer leukotrienes. If this were true, patients harboring the mutant genotype would likely not respond to antileukotriene treatment because their disease would be mediated by other factors.

This hypothesis was examined in a retrospective analysis of the response to an ALOX5 inhibitor, ABT-761, which has never been marketed but is clinically similar to zileuton. In 221 patients with asthma who received either high-dose ABT-761 (n = 114) or placebo (n = 107) treatment (36), the investigators found that ~ 6% of patients with asthma had no wild-type allele at the ALOX5 promoter locus and had a diminished response to ABT-761 treatment (Figure 2). These findings were consistent with the hypothesis that repeats of the -GGGCGG- sequence, other than the wild type, are associated with decreased gene transcription and ALOX5 product production. However, this aspect of the hypothesis has never been tested directly.

View larger version (16K): [in this window] [in a new window]   Figure 2.   ALOX5 genotype predicts antileukotriene response. Shown is the outcome of a clinical trial of ABT-761, an ALOX5 inhibitor similar to zileuton, stratified by genotype. Improvement in FEV1 from pretreatment baseline at 84 days of treatment was significantly greater for subjects possessing the wild-type (WT) genotype treated with ABT-761 (300 mg/day) compared with subjects possessing any ALOX5 mutant (Mut) allele. Modified from Drazen and coworkers (36).

The effects of this polymorphism on response to the cysteinyl-leukotriene antagonist zafirlukast have been reported in an abstract (37). In this study, patients without wild-type alleles at the ALOX5 promoter locus had a decrease in the FEV₁ in response to treatment with zafirlukast, whereas patients harboring at least one copy of the wild-type allele at this locus showed improvement in their FEV₁. These data confirm the findings with ABT-761 and lend credence to the yet unproven idea that this VNTR modifies treatment responses through modification of the synthesis of the leukotrienes.

Another enzyme of the leukotriene pathway, LTC₄ synthase, is responsible for the adduction of glutathione at the C-6 position of the arachidonic acid backbone (38). There is a known SNP in the LTC₄ synthase promoter, A444C, with a C allele frequency of 0.19 in normal subjects and 0.27 in patients with severe asthma (39). The 444C allele creates an activator protein-2-binding sequence that appears to be functional (40). These data suggested that the 444C variant is associated with enhanced cysteinyl-leukotriene production, and hence that patients with the A/A genotype may have leukotriene-driven asthma. Sampson and colleagues found that, among subjects with asthma treated with zafirlukast (20 mg twice daily), those homozygous for the A allele (n = 10 subjects) at the 444 locus had a lower FEV₁ response than those with the C/C or C/A genotype (n = 13). These findings provide possible evidence of a second pharmacogenetic locus in addition to the ALOX5 promoter locus modulating the leukotriene pathway. It will be of interest to determine whether individuals possessing multiple variant alleles at loci in the leukotriene pathway will have an additive response to treatment aimed at this pathway. Because the pathway is linear, our prediction is that any one DNA sequence variant associated with a decrease in cysteinyl-leukotriene synthesis will also be associated with a decreased response to therapy.

	STATISTICAL ISSUES

TOP INTRODUCTION PHARMACOGENETIC PATHWAYS AND... GENETIC ASSOCIATION ANALYSIS... PREVIOUS STUDIES OF ASTHMA... STATISTICAL ISSUES FUTURE DIRECTIONS AND ISSUES CONCLUSIONS REFERENCES

The current trend in genetic analysis of complex human diseases is away from family-based strategies using microsatellite markers and toward SNP genotyping and different analytical strategies based on association and haplotype analysis (9, 11, 41). Because response to asthma treatment varies with age, and the number and type of asthma medications are changing rapidly, it is unlikely that family-based asthma treatment data will be available in the foreseeable future. In the absence of these data, case-control association studies are the approach of choice. Case-control association analyses are now recognized as being well suited for localizing susceptibility loci (42), and they are intrinsically more powerful than linkage analyses in detecting weak genetic effects (41).

The testing of large numbers of SNPs for association with one or more traits raises important statistical issues regarding the appropriate false-positive rate of the tests and the level of statistical significance to be adopted given the multiple testing involved. The required methodological developments in genetic statistics are nontrivial, given the complexity of common diseases such as asthma (9).

The two major statistical issues in asthma pharmacogenetics relate to population stratification and statistical power.

Heterogeneity and Population Stratification

Phenotypic heterogeneity characterizes the response to asthma pharmacotherapy and can operate at two levels: between- individual heterogeneity and within-individual repeatability of response (4). Both are important factors to assess and consider in pharmacogenetic studies of asthma.

Genetic heterogeneity among different populations can also play an important role. In addition to variation in allele frequencies, there is also a high degree of variation in the strength of linkage disequilibrium in a given chromosomal region among populations of different origins (43) (each population is a product of its unique genetic history, including the timing of entry of new mutations into the population; this unique history may be reflected in the linkage disequilibrium patterns between SNPs in a given region) and also between different genomic regions (44, 45). Such genetic heterogeneity is a major challenge to gene discovery in asthma (2). Among the limitations of case-control association studies is the potential that undetected population stratification will produce misleading evidence of association.

Population stratification may cause spurious associations in a case-control study when allelic frequencies vary across subpopulations in a study cohort. For example, if there is an imbalance in ethnic group representation between the case and control cohorts, a spurious association could be detected (46). Such population stratification may result from recent admixture or from poorly matched cases and controls. Genotyping of random panels of SNPs, chosen without regard to the phenotype of interest, can be used to assure that case and control populations are genetically homogeneous. Methods have been developed to assess population stratification and, if necessary, to test correctly for association in the presence of such stratification (47-49). However, neither systematic testing for population stratification nor application of these new statistical methods has yet been incorporated into any pharmacogenetic studies, including studies of asthma.

Statistical Power

Growing experience with complex disease genetics has made clear the need to consider the issue of statistical power in genetic studies (41). Table 2 shows some simple estimation of required sample sizes of cases needed to detect a true odds ratio of 1.5 with 80% power and type I error probability () of either 0.05 or 0.005. Even for the "best case scenario"a common SNP acting in a dominant fashiona relatively large sample size of more than 300 cases (a total sample size of more than 900 subjects) is required at an  of 0.05 (Table 2).

Multiple testing issues are likely in many genetic association studies of candidate loci where either multiple SNPs in one gene, multiple SNPs in several loci, or both are tested, suggesting that an  of 0.005 is probably more realistic than an  of 0.05. Using the  of 0.005, or assuming an uncommon SNP (allele frequency less than or equal to 0.10) that acts in a recessive fashion, points to the need for large sample sizes, that is, more than 10,000 cases. Finally, Table 2 assumes an effect size (odds ratio of 1.5) that, in the context of a common multifactorial disease such as asthma, may be quite large. Assuming a smaller effect may be more realistic for many genes, and would lead to concomitantly higher required sample sizes. Simulation studies have also suggested that genes of small effect are not likely to be detectable by association studies in sample sizes of less than 500 (50).

Although these power calculations are simple and fairly conservative, they clearly demonstrate that the sample sizes used in many of the small case-control pharmacogenetic association studies conducted to date had insufficient power to detect even a large effect associated with an SNP. This suggests that larger scale studies than most of those currently being performed will be needed.

	FUTURE DIRECTIONS AND ISSUES

TOP INTRODUCTION PHARMACOGENETIC PATHWAYS AND... GENETIC ASSOCIATION ANALYSIS... PREVIOUS STUDIES OF ASTHMA... STATISTICAL ISSUES FUTURE DIRECTIONS AND ISSUES CONCLUSIONS REFERENCES

The ultimate goal of pharmacogenetics is to understand the role that sequence variation among individuals and populations plays in the variability of responses to pharmaceuticals. The frequency and penetrance of a sequence variant affecting responsiveness to a particular drug and potential interactions with other genetic and environmental factors must ultimately be assessed in multiple population-based samples. An SNP must be relatively common and have a significant impact on phenotype to be important at the population level in determining treatment response. These criteria become particularly important when extrapolating from specific clinical trials to general clinical use in the highly heterogeneous populations where asthma is most common and which are the current major markets for asthma therapeutics (3). Many extant clinical trials for asthma pharmacotherapy could be usefully expanded to include pharmacogenetic studies. Indeed, we contend that failure to archive DNA for pharmacogenetic analysis in a large asthma treatment trial would be a significant waste of resources. However, in the longer term it is clear that large, well-characterized cohort studies of population-based and ethnically diverse samples will be critical to the future success of any diagnostic SNP-based pharmacogenetic tests and for cost-effectiveness studies.

	CONCLUSIONS

TOP INTRODUCTION PHARMACOGENETIC PATHWAYS AND... GENETIC ASSOCIATION ANALYSIS... PREVIOUS STUDIES OF ASTHMA... STATISTICAL ISSUES FUTURE DIRECTIONS AND ISSUES CONCLUSIONS REFERENCES

Pharmacogenetic approaches to asthma offer great potential to improve our understanding and treatment of this disorder, but they also offer significant challenges. Although it is now possible to genotype patients with asthma at a few loci and to use this information to make therapeutic decisions that improve drug efficacy and mitigate complications, these studies have not yet been established in prospective clinical trials, and hence they cannot be adopted as the standard of care.

Current research in asthma pharmacogenetics has highlighted associations between SNPs in the -adrenergic receptors and modified response to regular inhaled -agonist treatments (e.g., albuterol). Variants within the 5-lipoxygenase gene have been suggested to predict the response to antileukotrienes in subjects with asthma. Confirmation of these findings, together with the current rapid creation of new knowledge, may mark the beginning of the clinical use of genotyping at an individual level as an adjunct to pharmacotherapy for asthma and many other disorders.

	Footnotes

Correspondence and requests for reprints should be addressed to Jeffrey M. Drazen, M.D., Pulmonary Division, Brigham and Women's Hospital and Harvard Medical School, 75 Francis Street, Boston, MA 02115. E-mail: jdrazen{at}nejm.org

(Received in original form September 25, 2001 and accepted in revised form January 18, 2002).

Acknowledgments: Supported in part by NIH grant 5 U01 HL65899-02 from the National Heart, Lung, and Blood Institute of the National Institutes of Health.

	References

TOP INTRODUCTION PHARMACOGENETIC PATHWAYS AND... GENETIC ASSOCIATION ANALYSIS... PREVIOUS STUDIES OF ASTHMA... STATISTICAL ISSUES FUTURE DIRECTIONS AND ISSUES CONCLUSIONS REFERENCES

1. Weiss KB, Sullivan SD. The health economics of asthma and rhinitis. I. Assessing the economic impact. J Allergy Clin Immunol 2001; 107: 3-8 [Medline].

2. Palmer LJ, Cookson WO. Genomic approaches to understanding asthma. Genome Res 2000; 10: 1280-1287 [Abstract/Free Full Text].

3. Hall IP. Pharmacogenetics of asthma. Eur Respir J 2000; 15: 449-451 [Medline].

4. Drazen JM, Silverman EK, Lee TH. Heterogeneity of therapeutic responses in asthma. Br Med Bull 2000; 56: 1054-1070 [Abstract/Free Full Text].

5. National Asthma Education Program. Guidelines for the diagnosis and treatment of asthma [summary of the Expert Panel II Report of the National Asthma Education and Prevention Program (NAEPP), National Heart, Lung, and Blood Institute (NHLBI), National Institutes of Health (NIH)].

6. Evans DAP, Manley KA, McKusick VA. Genetic control of isoniazid metabolism in man. Br Med J 1960; 2: 485-491 .

7. Marth GT, Korf I, Yandell MD, Yeh RT, Gu Z, Zakeri H, Stitziel NO, Hillier L, Kwok PY, Gish WR. A general approach to single-nucleotide polymorphism discovery. Nat Genet 1999; 23: 452-456 [Medline].

8. Gray IC, Campbell DA, Spurr NK. Single nucleotide polymorphisms as tools in human genetics. Hum Mol Genet 2000; 9: 2403-2408 [Abstract/Free Full Text].

9. Palmer LJ, Cookson WO. Using single nucleotide polymorphisms (SNPs) as a means to understanding the pathophysiology of asthma. Respir Res 2001; 2: 102-112 . [Medline]

10. Collins FS, Patrinos A, Jordan E, Chakravarti A, Gesteland R, Walters L. New goals for the US Human Genome Project: 1998-2003. Science 1998; 282: 682-689 [Abstract/Free Full Text].

11. Schork NJ, Fallin D, Lanchbury JS. Single nucleotide polymorphisms and the future of genetic epidemiology. Clin Genet 2000; 58: 250-264 [Medline].

12. Brookes AJ. The essence of SNPs. Gene 1999; 8: 177-186 .

13. Palmer LJ, Cookson WO. Atopy and asthma. In: Sham PC, Bishop T, editors. Genetic analysis of multifactorial diseases. London: BIOS Scientific Publishers; 2000. p. 215-237.

14. Silverman R. Treatment of acute asthma: a new look at the old and at the new. Clin Chest Med 2000; 21: 361-379 [Medline].

15. McGraw DW, Forbes SL, Kramer LA, Witte DP, Fortner CN, Paul RJ, Liggett SB. Transgenic overexpression of ₂-adrenergic receptors in airway smooth muscle alters myocyte function and ablates bronchial hyperreactivity. J Biol Chem 1999; 274: 32241-32247 [Abstract/Free Full Text].

16. Israel E, Drazen JM, Liggett SB, Boushey HA, Cherniack RM, Chinchilli VM, Cooper DM, Fahy JV, Fish JE, Ford JG, et al . The effect of polymorphisms of the ₂-adrenergic receptor on the response to regular use of albuterol in asthma. Am J Respir Crit Care Med 2000; 162: 75-80 [Abstract/Free Full Text].

17. Liggett SB. Pharmacogenetics of -1- and -2-adrenergic receptors. Pharmacology 2000; 61: 167-173 [Medline].

18. Liggett SB. The pharmacogenetics of ₂-adrenergic receptors: relevance to asthma. J Allergy Clin Immunol 2000; 105: S487-S492 [Medline].

19. Dewar J, Wheatley A, Venn A, Morrison J, Britton J, Hall I. ₂-Adrenoreceptor polymorphisms are in linkage disequilibrium but are not associated with asthma in an adult population. Clin Exp Allergy 1998; 28: 442-448 [Medline].

20. McGraw DW, Forbes SL, Kramer LA, Liggett SB. Polymorphisms of the 5' leader cistron of the human ₂-adrenergic receptor regulate receptor expression. J Clin Invest 1998; 102: 1927-1932 [Medline].

21. Green S, Turki J, Innis M, Liggett S. Amino-terminal polymorphisms of the human ₂-adrenergic receptor impart distinct agonist-promoted regulatory properties. Biochemistry 1994; 33: 9414-9419 [Medline].

22. Reihsaus E, Innis M, MacIntyre N, Liggett SB. Mutations in the gene encoding for the ₂-adrenergic receptor in normal and asthmatic subjects. Am J Respir Cell Mol Biol 1993; 8: 334-339 .

23. D'Amato M, Vitiani LR, Petrelli G, Ferrigno L, di Pietro A, Trezza R, Matricardi PM. Association of persistent bronchial hyperresponsiveness with ₂-adrenoceptor (ADRB2) haplotypes. A population study. Am J Respir Crit Care Med 1998; 158: 1968-1973 [Abstract/Free Full Text].

24. Barnes KC. Atopy and asthma genes: where do we stand? Allergy 2000; 55: 803-817 [Medline].

25. Hancox RJ, Sears MR, Taylor DR. Polymorphism of the ₂-adrenoceptor and the response to long-term ₂-agonist therapy in asthma. Eur Respir J 1998; 11: 589-593 [Abstract].

26. Tan S, Hall IP, Dewar J, Dow E, Lipworth B. Association between ₂-adrenoceptor polymorphism and susceptibility to bronchodilator desensitisation in moderately severe stable asthmatics. Lancet 1997; 350: 995-999 [Medline].

27. Drazen JM, Israel E, Boushey HA, Chinchilli VM, Fahy JV, Fish JE, Lazarus SC, Lemanske RF, Martin RJ, Peters SP, et al . Comparison of regularly scheduled with as-needed use of albuterol in mild asthma. Asthma Clinical Research Network. N Engl J Med 1996; 335: 841-847 [Abstract/Free Full Text].

28. Israel E, Drazen JM, Liggett SB, Boushey HA, Cherniack RM, Chinchilli VM, Cooper DM, Fahy JV, Fish JE, Ford JG, Kraft M, Kunselman S, Lazarus SC, Lemanske RF Jr,, Martin RJ, McLean DE, Peters SP, Silverman EK, Sorkness CA, Szefler SJ, Weiss ST, Yandava CN. Effect of polymorphism of the ₂-adrenergic receptor on response to regular use of albuterol in asthma. Int Arch Allergy Immunol 2001; 124: 183-186 [Medline].

29. Martinez FD, Graves PE, Baldini M, Solomon S, Erickson R. Association between genetic polymorphisms of the ₂-adrenoceptor and response to albuterol in children with and without a history of wheezing. J Clin Invest 1997; 100: 3184-3188 [Medline].

30. Drysdale CM, McGraw DW, Stack CB, Stephens JC, Judson RS, Nandabalan K, Arnold K, Ruano G, Liggett SB. Complex promoter and coding region ₂-adrenergic receptor haplotypes alter receptor expression and predict in vivo responsiveness. Proc Natl Acad Sci USA 2000; 97: 10483-10488 [Abstract/Free Full Text].

31. Samuelsson B, Dahlen SE, Lindgren JA, Rouzer CA, Serhan CN. Leukotrienes and lipoxins: structures, biosynthesis, and biological effects. Science 1987; 237: 1171-1176 [Abstract/Free Full Text].

32. Holgate ST, Bradding P, Sampson AP. Leukotriene antagonists and synthesis inhibitors: new directions in asthma therapy. J Allergy Clin Immunol 1996; 98: 1-13 [Medline].

33. Drazen JM, Israel E, O'Byrne PM. Treatment of asthma with drugs modifying the leukotriene pathway. N Engl J Med 1999; 340: 197-206 [Free Full Text].

34. In KH, Asano K, Beier D, Grobholz J, Finn PW, Silverman EK, Silverman ES, Collins T, Fischer AR, Keith TP, et al . Naturally occurring mutations in the human 5-lipoxygenase gene promoter that modify transcription factor binding and reporter gene transcription. J Clin Invest 1997; 99: 1130-1137 [Medline].

35. Silverman ES, Du J, De Sanctis GT, Radmark O, Samuelsson B, Drazen JM, Collins T. Egr-1 and Sp1 interact functionally with the 5-lipoxygenase promoter and its naturally occurring mutants. Am J Respir Cell Mol Biol 1998; 19: 316-323 [Abstract/Free Full Text].

36. Drazen JM, Yandava CN, Dube L, Szczerback N, Hippensteel R, Pillari A, Israel E, Schork N, Silverman ES, Katz DA, et al . Pharmacogenetic association between ALOX5 promoter genotype and the response to anti-asthma treatment. Nat Genet 1999; 22: 168-170 [Medline].

37. Anderson W, Kalberg C, Edwards L, Fling M, Binnie C, Nguyen Q, Sherman P, Sprankle K, Wagner M, Rickard K, Manasco P. Effects of polymorphisms in the promoter region of 5-lipoxygenase and LTC₄ synthase on the clinical response to zafirlukast and fluticasone. Eur Respir J 2000;16(Suppl B):183s.

38. Lam BK, Austen KF. Leukotriene C-4 synthase: a pivotal enzyme in the biosynthesis of the cysteinyl leukotrienes. Am J Respir Crit Care Med 2000; 161: S16-S19 [Free Full Text].

39. Sampson AP, Siddiqui S, Buchanan D, Howarth PH, Holgate ST, Holloway JW, Sayers I. Variant LTC₄ synthase allele modifies cysteinyl leukotriene synthesis in eosinophils and predicts clinical response to zafirlukast. Thorax 2000; 55: S28-S31 [Free Full Text].

40. Sanak M, Pierzchalska M, Bazan-Socha S, Szczeklik A. Enhanced expression of the leukotriene C₄ synthase due to overactive transcription of an allelic variant associated with aspirin-intolerant asthma. Am J Respir Cell Mol Biol 2000; 23: 290-296 [Abstract/Free Full Text].

41. Risch NJ. Searching for genetic determinants in the new millennium. Nature 2000; 405: 847-856 [Medline].

42. Silverman EK, Palmer LJ. Case-control association studies for the genetics of complex respiratory diseases. Am J Respir Cell Mol Biol 2000; 22: 645-648 [Free Full Text].

43. Zavattari P, Deidda E, Whalen M, Lampis R, Mulargia A, Loddo M, Eaves I, Mastio G, Todd JA, Cucca F. Major factors influencing linkage disequilibrium by analysis of different chromosome regions in distinct populations: demography, chromosome recombination frequency and selection. Hum Mol Genet 2000; 9: 2947-2957 [Abstract/Free Full Text].

44. Watkins WS, Zenger R, O'Brien E, Nyman D, Eriksson AW, Renlund M, Jorde LB. Linkage disequilibrium patterns vary with chromosomal location: a case study from the von Willebrand factor region. Am J Hum Genet 1994; 55: 348-355 [Medline].

45. Jorde LB, Watkins WS, Carlson M, Groden J, Albertsen H, Thliveris A, Leppert M. Linkage disequilibrium predicts physical distance in the adenomatous polyposis coli region. Am J Hum Genet 1994; 54: 884-898 [Medline].

46. Ewens W, Spielman R. The transmission/disequilibrium test: history, subdivision, and admixture. Am J Hum Genet 1995; 57: 455-464 [Medline].

47. Pritchard JK, Rosenberg NA. Use of unlinked genetic markers to detect population stratification in association studies. Am J Hum Genet 1999; 65: 220-228 [Medline].

48. Pritchard JK, Stephens M, Rosenberg NA, Donnelly P. Association mapping in structured populations. Am J Hum Genet 2000; 67: 170-181 [Medline].

49. Reich DE, Goldstein DB. Detecting association in a case-control study while correcting for population stratification. Genet Epidemiol 2001; 20: 4-16 [Medline].

50. Long AD, Langley CH. The power of association studies to detect the contribution of candidate genetic loci to variation in complex traits. Genome Res 1999; 9: 720-731 [Abstract/Free Full Text].