Intraocular Pressure Response to Topical β-Blockers Associated With an ADRB2 Single-Nucleotide Polymorphism FREE

Glaucoma is a progressive optic neuropathy¹ with an estimated prevalence of 1.86% in the US population aged 40 years and older.² Glaucoma has been called the “silent thief of sight,” and previous research has shown that half of the glaucoma cases in the community are undiagnosed.³ Timely diagnosis and treatment of glaucoma are necessary to prevent irreversible degeneration of retinal ganglion cells and concomitant vision loss. The goal of glaucoma management is to lower intraocular pressure (IOP); treatment options include pharmacologic agents, laser treatment, and surgery.⁴ Lowering IOP has been shown in clinical trials to reduce the risk of visual loss⁵ ; however, compliance in clinical settings has been low.^{6 - 7} There has been a recent trend away from the use of β-blockers as first-line medical therapy and toward newer therapies such as prostaglandin analogues.⁸ The result has been a concomitant rise in the overall cost of glaucoma drugs.⁹

Genetic variability, compliance, environment, and eye and systemic disease are thought to contribute to the overall IOP response to glaucoma medications.¹⁰ β-Blockers are used to treat elevated blood pressure and elevated IOP. Cytochrome P450 2D6 metabolizes timolol in vivo, and CYP2D6(GenBank AY545216) gene polymorphisms have been shown to be associated with timolol-related outcomes.¹¹ Topical β-blockers have been shown to have systemic effects,^{12 - 15} and outcomes related to systemic absorption of timolol are related to the CYP2D6genotype as well as antidepressants and other drugs known to inhibit CYP2D6.^{16 - 17} On the pharmacodynamic side, β₁- and β₂-adrenergic receptor genes are likely candidate genes for IOP response because they transduce the cellular effects of endogenous β-adrenergic agonists (eg, epinephrine).^{18 - 19} Also, systemic β-blockers have been shown to be associated with blood pressure response and mortality outcome following the systemic administration of β-blockers.²⁰

To determine whether candidate pharmacodynamic and pharmacokinetic gene polymorphisms are associated with IOP response to topical β-blockers, we genotyped 5 common, functionally relevant single-nucleotide polymorphisms (SNPs) in the β₁-, β₂-, and β₃-adrenergic receptor genes and 6 common, functionally relevant polymorphisms in the CYP2D6gene. We also genotyped 4 additional polymorphisms in 2 disease genes, optineurin and myocillin.

The electronic medical records of adults enrolled in the population-based Marshfield Clinic Personalized Medicine Research Project (PMRP)²¹ were searched from January 1960 to December 2005 to identify subjects with a diagnosis of ocular hypertension or glaucoma. The PMRP is a population-based biobank with stored DNA and serum samples for more than 19 000 subjects aged 18 years and older.²² All of the participants gave written, informed consent for the project, which includes access to Marshfield Clinic medical records for phenotyping. More than 95% of the residents within the geographic area selected for the PMRP use Marshfield Clinic for their health care, thus allowing for population-based epidemiologic research within the Marshfield Clinic system. Marshfield Clinic is an integrated regional health care system with 700 physicians in 41 locations serving approximately 360 000 patients throughout central and northern Wisconsin. All major medical specialties and subspecialties except whole-organ transplantation are covered within the clinic system. Except for the city of Marshfield, Marshfield Epidemiologic Study Area residents²³ reside rurally or in small towns or villages. The overall project and this substudy were approved by the Marshfield Clinic Institutional Review Board.

The medical records were manually abstracted for all IOP measurements, glaucoma diagnoses, and operations for glaucoma. The medical records were also manually abstracted for the concomitant use of systemic medications known to interact with topical β-blockers (ie, systemic β-blockers or selective serotonin reuptake inhibitors). Glaucoma diagnosis was confirmed with medical record evidence of 2 or more of the following: (1) glaucomatous visual field defect; (2) elevated IOP (> 21 mm Hg); (3) optic disc cupping (cup-disc ratio ≥ 0.8); or (4) rim narrowing characteristic of glaucoma. For this study, glaucoma suspects had only 1 of the preceding characteristics. Ten percent of the records were abstracted twice for quality assurance purposes. The IOP prior to topical β-blocker prescription and the lowest IOP in the first 3 months after prescription of a topical β-blocker were used to classify case or control status. Subjects were classified as cases for this study if their IOP in the first 3 months after being prescribed a topical β-blocker decreased by less than 20%. Subjects with IOPs that decreased by 20% or more were classified as control subjects.

Stored DNA samples were genotyped from the subjects who had used topical β-blockers and had baseline and follow-up IOP measurements within the first 3 months. TaqMan assays (Applied Biosystems, Foster City, California) were used. Validated assays were used for CYP2D6Cys188Thr, Gly1934Ala, and Cys2838Thr. Custom assays were designed by Applied Biosystems for the other 3 CYP2D6polymorphisms, Gly17949Cys, Ala2637del, and Gly4268Cys, necessary to assign a common haplotype. The CYP2D6genotypes were categorized functionally as extensive metabolizers (*1*1, *1*2, *2*2), intermediate metabolizers (*1*10, *1*3, *1*4, *2*10, *2*3, *2*4), and poor metabolizers (*3*4, *4*4).^{24 - 25}

Premade assays (Applied Biosystems) were used for the β₁-adrenergic receptor gene (ADRB1Ser49Gly and Arg389Gly). Custom-made assays (Applied Biosystems) were used for the β₂-adrenergic receptor gene (ADRB2Gly16Arg and Gln27Glu) and the β₃-adrenergic receptor gene (ADRB3Trp64Arg). Combinations of minor alleles for genotypes in the ADRB1, ADRB2and ADRB3SNPs were also calculated as suggested in a review of the pharmacogenetics of human β-adrenergic receptors.²⁶ For optineurin Glu50Lys and Met98Lys, assays from Applied Biosystems were used. Custom assays were developed for myocillin Gln368ter and −1000 Gly/Cys. Myocillin and optineurin were included because of their known genetic risk in certain forms of open-angle glaucoma. Test assays were set up; if there was a clear distinction between 11, 12, and 22 alleles, then the assay was run on all of the samples. The allele frequencies were compared with the known allele frequencies in dbSNP and were consistent with the known frequencies.

Data were entered twice and verified. For statistical analyses, SPSS version 15.0 statistical software (SPSS Inc, Chicago, Illinois) was used. To compare proportions, χ²analysis and Fisher exact test were used. Logistic regression was used to identify independent predictors of IOP response. The 95% confidence intervals (CIs) were calculated using the exact binomial distribution. P < .05 was considered statistically significant.

As of December 31, 2005, 18 773 adults were enrolled in the PMRP; all were included in this study. The overall rate of definite glaucoma in subjects aged 50 years and older was 2.07% (95% CI, 1.20%-2.38%) and the rate of treated ocular hypertension was 1.42% (95% CI, 1.19%-1.69%). Topical β-blockers were prescribed for 343 eyes of 215 PMRP subjects. Of these, 5 subjects receiving selective serotonin reuptake inhibitor medications at the same time they were receiving topical β-blockers were excluded from the study. Hence, a total of 210 subjects available for genotyping were included in this study. The sex distribution was 58.1% female (n = 122) and 41.9% male (n = 88). Their mean (SD) age on December 31, 2005, was 63.8 (11.3) years, ranging from 33.8 to 85.4 years. Of the subjects, 90 (42.9%) reported a family history of glaucoma on the initial questionnaire administered at the time of enrollment into the PMRP. Fifteen subjects (7.1%) were receiving systemic β-blockers at the same time they were receiving topical β-blockers. All of the polymorphisms were found to be in Hardy-Weinberg equilibrium. The frequency of alleles defining the CYP2D6haplotype were as follows: 40.1% for *1, 35.2% for *2, 3.3% for *3, 19.5% for *4, and 1.9% for *10, which is similar to what has been reported previously in white subjects.^{23 - 24} Of the 210 subjects genotyped, 28 (13.3%) had allele combinations that could not be assigned to common CYP2D6haplotypes.

The mean (SD) IOP in the right eye was 24.9 (5.9) mm Hg at baseline and 19.1 (3.9) mm Hg at follow-up. The mean (SD) IOP in the left eye was 24.8 (5.9) mm Hg at baseline and 18.8 (3.6) mm Hg at follow-up. The greatest change in IOP in either eye in the first 3 months ranged from −70.8 mm Hg to +25.0 mm Hg (median change, −23.3 mm Hg). An IOP reduction of 20% or more in the first 3 months was observed for 91 right eyes (55.2%) and 92 left eyes (54.4%). An IOP reduction of 20% or more in the first 3 months in 1 or both eyes was observed in 128 subjects (61.0%). Subjects treated with β-blockers alone had significantly higher response rates than subjects treated with a combination of β-blockers and other IOP-lowering medications (70.3% vs 40.1%, respectively; χ² = 14.7; P < .001). In this drug-exposed study cohort (n = 210), neither age (ttest = 1.48; P = .14) nor use of systemic β-blockers (χ² = 0.01; P = .94) was related to IOP response. Men were significantly more likely than women to have either 1 or both eyes respond with an IOP decrease of 20% or more with topical β-blockers (69.3% vs 54.9%, respectively; χ² = 4.48; P = .04).

Tables 1, 2, and 3display the comparison of genotype distribution for CYP2D6, ADRB, and glaucoma disease genes, respectively, between subjects who did or did not have an IOP decrease of 20% or more. Variables were created to combine genotypes. Neither predicted CYP2D6phenotypes (Table 1) nor optineurin or myocillin gene polymorphisms (Table 3) were related to IOP response. Optineurin Glu50Lys was not polymorphic in this population. The ADRB-coding SNPs were not associated with IOP-lowering efficacy in univariate analyses (Table 2). Variables that combined the minor alleles for the 5 ADRBSNPs were created. None were found to be statistically significant (data not shown).

Table 4shows the results of the logistic regression models for ADRB2, developed to predict an IOP decrease of 20% or more with β-blocker use and adjusted for sex, family history of glaucoma, and use of systemic β-blockers. For both ADRB1and ADRB3, there were too few subjects homozygous for the minor allele to allow for multivariate logistic regression analyses. After adjusting for sex, family history of glaucoma, and use of systemic β-blockers, subjects with the homozygous major allele (CC) genotype for rs1042714 were significantly more likely than subjects with the heterozygous (CG) genotype to experience an IOP decrease of 20% or more (odds ratio, 2.00; 95% CI, 1.00-4.02).

In this study, we found that candidate pharmacodynamic but not pharmacokinetic polymorphisms were associated with IOP response to topical β-blockers. Specifically, a coding SNP in the ADRB2gene (Gln27Glu) was associated with 2-fold greater odds of a clinically meaningful reduction in IOP following treatment with a topical β-blocker.

The β-adrenergic receptor is a member of the adrenergic family of G-protein–coupled receptors. Epinephrine and norepinephrine are the primary endogenous agonists, but other endogenous catecholamines (eg, dopamine) can interact with these receptors as well. In the early 1990s, investigators characterized several ADRB2polymorphisms with altered signaling properties.^{27 - 29} In 1999, the same group reported an ADRB1polymorphism that altered the cytoplasmic tail near the seventh transmembrane-spanning segment.³⁰ Using site-directed mutagenesis, these investigators were able to show that the resulting amino acid change (Gly389Arg) in ADRB1was associated with differential adenylate cyclase activation in permanently transfected fibroblasts (CHW-1102 cells). Other investigators have linked this ADRB1SNP to altered receptor expression.³¹ Two nonsynonymous coding SNPs in ADRB2have also been associated with altered cellular receptor trafficking (in vitro).³² Importantly, these SNPs have also been associated (in vivo) with altered clinical outcomes, including asthma³³ and acute coronary syndromes.²⁰ Further, there appears to be marked linkage disequilibrium between these 2 ADRB2SNPs,³⁴ and future research may benefit from the consideration of diplotypes.

Three other studies have looked at the β₂-adrenergic receptor gene in relation to glaucoma and IOP, with varying results. In a study³⁵ of healthy volunteers, a single candidate β₂-adrenergic receptor polymorphism was found to not be associated with IOP response to topical timolol. The differing results between our study and the previous study of the β₂-adrenergic receptor gene and IOP could be owing to the differing study populations or a lack of statistical power in the previous study. In a study³⁶ of 583 subjects, the β₂-adrenergic receptor gene was not found to be a glaucoma susceptibility locus. In a study³⁷ of 505 Japanese subjects, the IOP at glaucoma diagnosis was found to be significantly higher in patients carrying 27Glu. This coding SNP (C79G transversion) induces a nonconservative amino acid substitution (Gln27Glu) near the N terminus of the ADRB2gene product.³³ The result is an alteration in agonist-promoted downregulation of the receptor.²⁸ In vitro studies have demonstrated that Gln27Glu affects receptor function.³⁸ A 60-fold greater isoprenaline concentration was required to downregulate Glu27 to the same extent as Gln27.

One of the major strengths of this study is the population-based nature of the study cohort, which allows inferences to the entire population to be made. Limitations of the study include the fact that the potential effect of noncompliance on the study results cannot be determined and the fact that IOP data were not available for all of the subjects, limiting statistical power for some analyses. Prospectively collected data with a standardized protocol and a larger study cohort could lead to statistically significant results for some of the other SNPs that were of borderline significance.

In conclusion, we found that a coding SNP in the ADRB2gene is associated with an increased likelihood of a clinically meaningful IOP response to topical β-blockers. Topical β-blockers are the least expensive agents used to treat glaucoma and ocular hypertension; therefore, the potential for genotype-based prescribing could save health care dollars. These findings need to be confirmed in other populations and evaluated prospectively.

Correspondence:Catherine A. McCarty, PhD, MPH, Center for Human Genetics, Marshfield Clinic Research Foundation, 1000 N Oak Ave, Marshfield, WI 54449 (mccarty.catherine@mcrf.mfldclin.edu).

Submitted for Publication:November 1, 2007; final revision received December 19, 2007; accepted December 21, 2007.

Author Contributions:Dr McCarty had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Financial Disclosure:None reported.

Funding/Support:This study was funded in part by a grant from National Glaucoma Research, a program of the American Health Assistance Foundation.

Additional Contributions:Terrie Kitchner and William Hubbard, BS, provided medical record abstraction, Carla Rottscheit provided data management, and Kai Qi Zhang, BS, provided genotyping. Marshfield Clinic Research Foundation supported this work through the assistance of Linda Weis and Alice Stargardt in the preparation of the manuscript.