| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Division of Therapeutics and Molecular Medicine, University Hospital of Nottingham, Nottingham, UK.
Correspondence to: Ian P. Hall, DM, Division of Therapeutics and Molecular Medicine, University Hospital of Nottingham, Nottingham NG7 2UH, UK; e-mail: Ian.Hall{at}nottingham.ac.uk
Abstract
Pharmacogenetics offers the potential to optimize treatment for individual patients by using genetic information to improve efficacy or avoid side effects. While there are a number of examples in which the approach is already in routine clinical usage, exploitation of this approach in asthma is still under development. A number of examples of possible pharmacogenetic approaches that may prove of value in the management of asthma are discussed.
Key Words: asthma • ß2-agonists • pharmacogenetics
The response of an individual patient to any given drug depends on a range of factors, including compliance, comorbidity, disease severity, and genetic background. The great hope for pharmacogenetics is that it will provide sufficient power to predict either treatment response (efficacy) or the risk of adverse drug reactions (ADRs) in the general population and that it will prove cost-effective to genotype individuals before treatment. While this hope may seem overambitious at first glance, there are already a number of good examples of pharmacogenetic approaches to disease management that form part of routine clinical care. One situation in which pharmacogenetic approaches are already used is in thrombophilia screening before determining regimens for anticoagulation: individuals with factor V von Lieden (or other genetically determined thrombophilias) will be treated differently than those without underlying genetic risk factors. The number of clinical situations in which genotyping is performed before treatment will almost certainly increase as licensing bodies approve drugs for use in specific subsets of patients with disease. Another potential influence on the incorporation of pharmacogenetics into routine care is through guideline development, an example here being the recommendation by some disease-based guidelines for thiopurine methyl transferase genotyping before commencing azathioprine therapy. Pharmacogenetics is also used in oncology, in which tumor profiling helps determine the most appropriate use of chemotherapeutic agents. It is only a matter of time before these approaches spill over into mainstream prescribing.
In order to determine the potential contribution of genetic approaches to predicting treatment response for any given drug, a number of key questions need to be asked. These are as follows: (1) is there genetic variability either in the primary drug target, its downstream effector pathways, or key enzymes/transporters that determine pharmacokinetics; (2) if genetic variability exists in any of these targets, are these genetic variants of functional significance; (3) if functionally significant variants can be identified, what proportion of the overall efficacy or risk of ADRs can be predicted by these variants; and (4) would it be cost-effective to screen for these variants within the population before prescribing?
Most of the drugs used in the management of asthma do not have narrow therapeutic windows (with the exception of theophylline), and hence to date the majority of work has concentrated on pharmacodynamic rather than pharmacokinetic variables.1 Table 1 gives a list of the major airway drug targets and the state of polymorphism screening as it stands at present. It is clear that the coding regions for the major drug targets have been extensively screened at least in white populations, but less is known about the regulatory elements controlling gene expression of these targets. This is likely to be important because the effectiveness of a drug depends not only on the direct interaction of that drug with a target but also on the level of expression of the target. Table 1 also shows that there are relatively few targets for which good clinical data exist. However, a number of studies have either recently been completed or are ongoing that will fully determine the likely contribution of pharmacogenetic variability in these major drug targets (or other genes) to treatment response.
|
The best-studied example of the potential contribution of pharmacogenetics to treatment response in asthma comes from work on the human ß2-adrenoceptor (Fig 1 ). While in many ways this provides potential proof of concept for this approach, studies have also highlighted the potential difficulties. The ß2-adrenoceptor is highly polymorphic,2 with three functionally relevant coding region polymorphisms (Arg 16 Gly, Gln 27 Glu, Thr 164 Ile) and multiple single-nucleotide polymorphisms (SNPs) either elsewhere in the coding region of the gene, in the 5' untranslated region or the 3' untranslated region of the gene, or in the adjacent genomic region. Functional data suggest that the Ile 164 variant has reduced coupling to adenyl cyclase and that the time course of binding of some ligands such as salmeterol to this variant of the receptor is reduced, resulting in a shorter duration of receptor activation.34 In contrast, the N-terminal polymorphisms at codon 16 and codon 27 do not produce clear differences in either ligand binding or basal activation of adenyl cylclase but do alter downregulation profiles with enhanced downregulation being seen at least in transformed cell systems with the Gly16 variant.5 The majority of functional data have been generated using either recombinant cell systems or, for noncoding region polymorphisms, using transfection approaches. The difficulty with this approach is that the SNPs are studied in isolation out of context of the other genetic variability in the nearby genomic region. Given that > 50 SNPs have now been described within close proximity of the human ß2-adrenoceptor gene, it has therefore been argued that in vivo it is the combination of genotypes (or haplotype) across this region that is important.6 In practice, however, if extended haplotypes involving multiple SNPs are to be studied, the number of individuals required for clinical studies becomes unfeasible. A number of investigators have therefore chosen to study limited-framework haplotypes that account for the majority of the variation at the locus. The difficulty with this approach is that the common haplotypes may be different in different racial groups and that investigators have often chosen different SNPs to define their haplotypes, which makes comparison between studies difficult.
|
Cys Leukotriene Pathway Pharmacogenetics
The other major class of asthma drugs for which extensive data exists on the potential contribution of pharmacogenetic variability to treatment response is the Cys leukotriene receptor antagonists. Initial attention on the pathway (Fig 2 ) focused on genetic variability in the 5 lipoxygenase (5-LO) [ALOX 5] gene as a potential determinant to response to 5-LO inhibitors.13 In one early clinical study,14 the investigators were fortunate enough to be able to genotype subjects from a phase 3 clinical trial that examined a novel 5-LO inhibitor (ABT 761) in subjects with mild-to-moderate asthma. This study demonstrated that individuals who had non-wild type repeats of a transcription factor binding motif in the ALOX 5 gene promoter had reduced clinical responses to 5-LO inhibition.14 With renewed interest in the possible use of 5LO inhibitors in the treatment of asthma, further studies may be required on this polymorphism. In contrast, no clear effect of this polymorphism was seen (at least in heterozygous individuals) on responses to Cys leukotriene receptor antagonists.15 However, there are multiple polymorphisms in other key regulatory enzymes responsible for synthesis of the important bioactive airway leukotrienes. The majority of attention has focused on a promoter polymorphism in leukotriene C4 synthase, which appears to predict at least in part responses to leukotriene receptor antagonists.1617 However, logically one might predict that it will be the combination of the polymorphisms in these different key regulatory enzymes and receptors that may ultimately determine treatment response. There have been some attempts to tease out the possible contribution of different genes important in this pathway for treatment response to a Cys leukotriene receptor 1 antagonist.18 However, because of the number of potential gene variants that may contribute to efficacy, large studies will be needed to fully evaluate the potential contribution of pharmacogenetic variability in this pathway to treatment response to Cys leukotriene receptor 1 antagonists. Work in the cardiovascular field has demonstrated the potential importance of genetic variants in this pathway to disease risk and also to treatment response,19 suggesting the potential for important effects to be defined in asthma.
|
The third major pathway that has been investigated for genetic variability that might predict treatment response is the pathways underlying response to glucocorticoids. Two approaches have been utilized here. In one approach, a true pharmacogenomic approach led to the identification of a number of key genes whose expression profiles appear to predict steroid responsiveness in in vitro assays utilizing peripheral blood mononuclear cells.20 The alternative has been a more traditional SNP-based pharmacogenetic approach that identified the gene for CRHR1 as a potential marker of steroid response.21 However, definitive studies are required to fully determine the likely predictive value of polymorphisms in these genes or of expression profiling as potential clinical markers of steroid responsiveness. One difficulty here is adequate clinical phenotyping of steroid responsiveness. The other important issue to deal with is whether or not there is dissociation between effects on efficacy and on steroid-related side effects such as osteoporosis: if specific risk factors could be identified that predicted long-term side effects from steroid usage, then informed decision making on treatment strategies (eg, the use of alternative immunosuppression) could be made.
Genome-Wide Association
The above examples have dealt in general with the use of intensive sequencing of key candidate genes to identify potential polymorphic variants that a priori might be predicted to be important in determining treatment response. However, it is likely that for many targets many other gene products can also potentially predict treatment response. Where the number of genetic variables that account for pharmacogenetic effects for a given treatment response is very large, it seems unlikely that wide-scale genotyping will become useful in the clinical setting. However, for some genes at least it is much more likely that this approach will be successful, and encouragingly examples from other disease areas suggest that genotyping for relatively small numbers of variants may provide adequate information; a good example is the ability of a combination of cytochrome P450 2C9 status and vitamin K epoxide reductase status to predict anticoagulation efficacy with warfarin. When it proves necessary to search for multiple potential contributing genetic factors, this approach will be markedly facilitated by the advances made in understanding haplotype structure across the human genome as a consequence of the Hap Map project. In addition, the potential to genotype large numbers of SNP in individuals at relatively low cost using either chip- or bead-based platforms will lead to a radical change in our approach to pharmacogenetics. The key factor that will limit this approach will be access to good data on treatment outcomes in well-characterized populations of an adequate size.
Target Discovery
Over the last 10 years, there have been a number of genome-wide linkage studies that have attempted to discover novel gene variants that might contribute to the risk of developing asthma. This has led to the identification of several genes (including ADAM 33, DPP10, PHF 11 and GPR 154) that might potentially contribute to asthma pathogenesis, and that might therefore also represent novel therapeutic targets. However, relatively little is known to date about the way in which functional variants in these genes might contribute to asthma pathophysiology, and hence at presence the potential to target the products of these genes for novel therapeutic agents remains unclear.
Conclusions
Pharmacogenetic approaches have been developed using information derived from the investments made in the human genome project and subsequently the SNP and Hap Map projects. The attraction of pharmacogenetics is that it provides an obvious way to use genetic information to improve patient care. However, in order for this approach to be clinically useful it requires pharmacogenetic factors to have at least reasonable predictive value either for efficacy or ADRs for given treatments, and also it needs to be shown that this approach is cost-effective in routine clinical care. This is most likely to be the case for guiding the use of expensive drugs, in situations in which efficacy is only apparent after prolonged treatment, or for the prevention of ADRs. Despite these reservations, there are already clear examples where pharmacogenetic approaches are proving valuable. It is likely therefore that these approaches will gradually be introduced into the management of patients with a range of respiratory diseases over the next 10 years.
Footnotes
Abbreviations: ADR = adverse drug reaction; 5-LO = 5 lipoxygenase; LABA = long-acting ß2-agonist; SNP = single-nucleotide polymorphism
The author has no conflicts of interest to disclose.
Received for publication May 30, 2006. Accepted for publication September 15, 2006.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
