Pharmacogenetics of β2-agonists in children

β2-Agonists have improved lung function and quality of life in many patients with asthma. Together with steroids, they have been considered a valuable tool to fight asthma, a disease affecting millions of adults and children around the world. Relieving shortness of breath and airway obstruction has been viewed as a milestone in asthma therapy by many patients and doctors. The introduction of long-acting β2-agonists (LABAs) more than 15 years ago increased the quality of life for patients with asthma. When taken together with inhaled corticosteroids, their therapeutic value is evident.

However, not all patients with asthma benefit from LABA treatment. Some may even become worse and experience an increase in severe asthma exacerbations. These concerns were addressed in the Salmeterol Multicenter Asthma Research Trial (SMART) study, and after the analysis of the first 26,000 subjects in the trial, the study was discontinued because of a 4-fold increase in asthma-related mortality in the salmeterol group over the placebo group.1 These results led the US Food and Drug Administration to issue a public health advisory in 2006 stating that “these medicines (LABA) may increase the chance of severe asthma episodes, and death when those episodes occur.”2 A large meta-analysis of formoterol effects including approximately 68,000 patients published only recently indicated a 50% increased (but nonsignificant) risk for asthma-related death in the formoterol treatment group.3 Most recent national and international guidelines have already addressed the issue. The use of LABA without concomitant inhaled corticosteroid treatment is strongly discouraged. A number of commentaries and editorials have already discussed the pros and cons of LABA use extensively, and the US Food and Drug Administration readdressed the risks and benefits of LABA use in children in a December 2008 meeting.4 It is a common request of the scientific community that more data and well designed studies on the subject are needed fast and that the pharmaceutical industry has to provide these studies irrespective of costs.5, 6 There is an urgent need to understand the mechanisms underlying the associations among LABA use, severe exacerbations, and death. It is important to study every asthma-related death in connection with LABA to learn what went wrong in those cases and to identify individual risk factors for these adverse effects.

Genetic predisposition may contribute to the effect of β2-agonists. The β2-adrenergic receptor gene (ADRB2) coding for the protein that is the target of β2-agonists is highly polymorphic, harboring a number of common and rare genetic variations already identified. Patients who are arginine homozygotes (Arg/Arg) at amino acid position 16 of the β2-adrenergic receptor seem to have an increased risk for asthma exacerbations and a deterioration in lung function compared with carriers of the other 2 possible genotypes (Arg/Gly and Gly/Gly) when short-acting β2-adrenergic receptors were used for a longer period.7, 8, 9 Existing data on the effect of this polymorphism in combination with LABA use is controversial, but data are limited.10

The study by Basu et al11 in this issue of the Journal studies the effects of the common ADRB2 Gly/Arg 16 polymorphism in combination with short-acting and/or long-acting β-adrenergic agonists on asthma exacerbation in a large population of children. The study has shortcomings, mainly because it is a retrospective observation study, and thus a reference population of patients with moderate to severe asthma not exposed to LABAs or short-acting β2-agonists does not exist in the study—a shortcoming that is acknowledged by the authors and not unexpected on the basis of the widespread use of these medications in clinical practice. Taking this into account, the study results are still disturbing in that clear and significant effects are observed: the Arg16 allele significantly increases the exacerbation risk of children and young adults with asthma when they are regularly exposed to short-acting β2-agonists and/or LABAs. These results indicate that short-acting β2-agonists and LABAs may have similar effects in genetically susceptible individuals and that regular short-acting β2-agonists on demand may not be an alternative to LABAs in these patients. The results strengthen the argument that genetic information is necessary and important to interpret the existing datasets on LABA in the ongoing discussion. Any future pharmacologic study on β-adrenergic agonists (or retrospective analysis thereof) not addressing this aspect can be considered flawed.

The authors are aware of the limitations of their study, namely that they cannot address the hypothesis that the use of non-β2 relievers is not associated with the same effects that have been reported with inhaled β2-agonist relievers. Rightly, they conclude that proper studies involving a non-β2 reliever are now required, and they propose 2 strategies. First, a population of patients with asthma with significant use of non-β2 relievers could be studied to compare the interactions of the Arg16 genotype with inhaled β2-agonist reliever use versus non–β2-agonist reliever use. Second, a population of patients with asthma carrying the Arg16 genotype could be prospectively randomized to inhaled β2-agonist versus non–β2-agonist reliever therapy. If these future studies support the hypothesis that the Arg16 allele interacts with β2-agonist use to increase exacerbations, it may have severe implications on clinical practice. Genetic testing may become necessary to minimize individual risk when administering β2 relievers, especially if we do not find other ways to identify those that are at risk when inhaling β2 relievers. Is this feasible, and will it be really necessary? The answer is yes to both. “Do no harm” is a central part of medical ethics, and ignoring information that can help to minimize side effects is unethical. However, we should have firm evidence for such interventions.

It becomes clear here that pharmacogenetics in the early 21st century will not primarily lead to the invention of individualized therapeutics but exclude certain groups of patients from established treatment regimens because of their genetic risk profile. This should (in theory) make treatment safer for the rest of the patients and initiate further drug developments. This is a general observation not restricted to the β2-agonist discussion. With the millions of polymorphisms we all are carrying, everybody will have personal risk profiles for drug interactions. This information becomes available rapidly. We need to start dealing with it in clinical practice whether we like it or not.