Methods

Patients with heart failure (n = 93) who had undergone the initiation and titration of their β-blocker therapy in the Mayo Clinic Heart Failure Clinic (Rochester, Minnesota), were recruited for the present study. The clinical charts were reviewed to identify patients meeting the inclusion and exclusion criteria. The inclusion criteria were age >18 years, chronic heart failure of ≥6 months’ duration, New York Heart Association functional class I to IV, an ischemic or a nonischemic etiology of heart failure, and left ventricular ejection fraction of ≤45%, measured within 1 year of study enrollment. Criteria were also developed prospectively to characterize this retrospective cohort as responders or nonresponders. The response to β-blocker (carvedilol or metoprolol) therapy was classified according to the following criteria. The duration of the drug titration regimen that was required to achieve the guideline recommended or maximally tolerated β-blocker dose compared to the interval of the standard protocol-driven dose titration (β-blocker doses were initiated at a low dose [carvedilol 3.125 mg twice daily; metoprolol tartrate 12.5 mg twice daily] and titrated by doubling the dose every 10 days until the goal dose or highest symptom-limited dose was achieved). Second, the achievement and toleration without exacerbation of heart failure symptoms of the highest tolerated dose (i.e., dose regimen sustained with a goal dose for carvedilol of 25 mg twice daily if the body weight was ≤85 kg and 50 mg twice daily if the body weight was >85 kg and a goal dose for metoprolol tartrate of 100 mg twice daily). Third, an increase of one classification in the New York Heart Association functional status. Fourth, an increase in the left ventricular ejection fraction (of ≥10%) by echocardiographic evaluation. Finally, a ≥10% increase in the distance a patient could walk in the 6-minute walking test. Of these 5 criteria, 3 were needed to be considered a responder to β-blocker therapy. The nurse and physician who evaluated responder status were unaware of the genotyping information. The Mayo Foundation institutional review board approved the study, and all patients provided written informed consent for study participation. Genomic DNA was extracted from whole blood specimens for all genotyping assays using a Qiagen BioRobot EZ1 extraction platform (Valencia, California), according to the manufacturer’s recommendations.

Genotyping of ADRB1 Arg389Gly was performed using polymerase chain reaction (PCR) amplification and multiplexed allele-specific primer extension (ASPE) incorporating biotinylated dCTP. The primers and cycling conditions for the PCR and ASPE are available on request. The beads were then hybridized to the universal tag sequence at the 5′ end of the allele-specific primers. The biotinylated dCTP in the allele-specific products was conjugated with streptavidin and a phycoerythrin reporter. Positive signals were sorted and read by a Luminex 100 xMAP instrument (Toronto, Ontario, Canada). Raw median fluorescence intensity signals were analyzed, and genotyping calls were made according to the allelic ratios.

Genotyping of CYP2D6 was performed with a Luminex xTAG kit, incorporating a multiplex PCR and ASPE with proprietary Universal Tag sorting system on the Luminex 100 xMAP platform, as described in the package insert. Variations detected by this kit include 1584C>G, 100C>T, 124G>A, 883G>C, 1023C>T, 1707T>del, 1758G>T, 1846G>A, 2549A>del, 2613delAGA, 2850C>T, and 2935A>C. In brief, 2 multiplex PCRs were performed and then pooled together and subjected to the ASPE reaction. The specific conditions used for PCR and ASPE are available on request. The biotinylated dCTP in the ASPE products was conjugated with streptavidin and a phycoerythrin reporter. Positive signals were sorted and read by a Luminex100 xMAP instrument. Genotypes were assigned using the xTAG Data Analysis software. According to the characteristics of the CYP2D6 alleles present, genotypes were categorized as ultrarapid metabolizers (one or more duplicated allele or one or more allele with increased activity), extensive metabolizers (EMs; 2 wild-type alleles), intermediate-to-extensive metabolizers (one wild-type and partially defective allele), intermediate metabolizers (IMs; one wild-type and one null allele or one null plus one partially defective allele), or poor metabolizers (PMs; 2 null alleles).

Genotyping of the UGT1A1 promoter TA repeat polymorphism was performed using a PCR protocol that included a 6FAM fluorescent dye labeled forward (Applied Biosystems, Foster City, California) and reverse 5′-GTTTCTT-3′ Tailed (Idaho Technology, Salt Lake City, Utah) primer mix. The PCR conditions are available on request. The amplicons were then diluted 1:10 with DNase/RNase-free water before combining the diluted DNA (1 μL) with formamide (12.5 μL; Applied Biosystems) and a 500-bp ROX size standard (0.5 μL, Applied Biosystems). The samples were separated using capillary electrophoresis on the ABI PRISM 3100 Genetic Analyzer with Performance Optimized Polymer 4. After the run, sample migration distances were analyzed using the Genotyper software (Applied Biosystems) to determine the genotype.

The relations of the categorical clinical characteristics with response to treatment were assessed using the Fisher exact test. The relations of the continuous clinical characteristics with response to treatment were assessed using the Wilcoxon rank sum test, because several of the laboratory measurements had skewed distributions. Associations between the ADRB1 , CYP2D6 , and UGT1A1 genotypes (or corresponding enzymatic phenotypes) with the response (no vs yes) to medication were assessed using logistic regression an analysis and indicator variables to represent the alleles/phenotypes of interest. The relation between the dosage of β-blocker medication used and the ADRB1 , CYP2D6 , and UGT1A1 genotypes/phenotypes were assessed using linear regression analysis, with indicator variables to represent the alleles of interest. Significance was defined as p <0.05.

Results

The study population consisted of 93 patients with heart failure. The mean ± SD age was 65.2 ± 11.5 years, and 25.8% were women ( Table 1 ). Of the 93 patients, 19 were taking metoprolol tartrate and 74 were taking carvedilol.

The patients were enrolled and the samples analyzed during a 14-month period in 2007 to 2008. According to the response criteria defined, the patients were classified retrospectively as nonresponders or responders to metoprolol or carvedilol. The principal features that characterized nonresponders were an inability to tolerate standard dose titration and no improvement in the left ventricular ejection fraction at the follow-up clinical evaluations. Nonresponders typically required 2 to 3 times the standard titration period to achieve the maximally tolerated or goal dose of β blocker owing to side effects such as worsening fatigue, dizziness, or worsening dyspnea with exertion. Nearly all responders had an increase in the left ventricular ejection fraction. Of the 19 patients treated with metoprolol, 10 (52.6%) were nonresponders. Of the 74 patients treated with carvedilol, 32 (43.2%) were nonresponders. No significant difference was found between gender, age, and chronic obstructive pulmonary disease and the response to either drug (p = 0.38, p = 0.26, and p = 0.14, respectively). However, an association was found between diabetes mellitus and the overall response (odds ratio 0.30, 95% confidence interval 0.12 to 0.74; p = 0.009) and the response to carvedilol (odds ratio 0.24, 95% confidence interval 0.08 to 0.66; p = 0.006) but not metoprolol (p = 0.70). The Fisher exact analysis demonstrated no relation between the frequency of response and the specific drugs (p = 0.61).

The observed genotype distributions ( Table 2 ) were within Hardy-Weinberg equilibrium (p >0.29 for all) and were similar to the reported frequencies for patients of white ancestry. The allelic and/or phenotypic frequencies of ADRB1 , CYP2D6 , and UGT1A1 variants did not differ significantly between responders and nonresponders for metoprolol or carvedilol (p ≥0.18 for all; Table 2 ). Using the Fisher exact test, no significant difference was found on the univariate analyses of ADRB1 genotype (individual genotype [Arg/Arg, Arg/Gly, or Gly/Gly] or ADRB1 389Gly carrier [Arg/Gly and Gly/Gly]) and the response to metoprolol (p >0.18 for all). Also, no significant difference was found between the CYP2D6 phenotype (ultrarapid metabolizer, EM, intermediate-to-extensive metabolizer, IM, or PM) and response to metoprolol (p = 0.75). The CYP2D6 phenotypes were divided into individual and subgroups and compared against each other (PM vs all other phenotypes together; ultrarapid metabolizers vs EM vs intermediate-to-extensive metabolizer + IM vs PM; and ultrarapid metabolizers vs EM + intermediate-to-extensive metabolizer + IM vs PM). Whichever group included EM was considered the wild-type group. No statistically significant difference was found in the response versus nonresponse to metoprolol for any of the CYP2D6 analyses (p >0.55 for all). The UGT1A1 phenotypes were also divided into individual and subgroups and compared against each other (PM vs IM vs EM; PM vs IM + EM; EM vs IM + PM). UGT1A1 was not associated with the response to metoprolol (p >0.21 for all). In examining the response to carvedilol, the same analyses involving ADRB1 and CYP2D6 also held true for carvedilol. Additionally, no significant difference was found between the response to carvedilol and the UGT1A1 phenotype (EM, IM, or PM or grouped as EM vs IM + PM or EM + IM vs PM; p >0.21 for all).

The median dose for metoprolol responders and nonresponders was 150 mg/day (range 50 to 200) and 100 mg/day (range 12.5 to 200), respectively. The median dose for carvedilol responders and nonresponders was 50 mg/day (range 20 to 100 and 25 to 100, respectively). The relation between the dose of β blocker and genotype was analyzed. The data listed in Table 3 show that weak univariate relations were found between the β-blocker dosage and ADRB1 genotypes and CYP2D6 phenotypes, but not UGT1A1 or ADRB1 389Gly carriers. In particular, those patients with the Gly/Gly ADRB1 genotype required about 24.94 mg/day more carvedilol than those with the wild type (model 2; p = 0.01). No difference was seen in the metoprolol dosage according to the ADRB1 genotype. Similarly, those patients with the PM CYP2D6 phenotype required about 19.79 mg/day more carvedilol than those with any other CYP2D6 phenotype (model 4; p = 0.02). No difference was seen in the metoprolol dosage according to CYP2D6 phenotype.

Table 3

Univariate relations between dosage of β blockers and ADRB1, CYP2D6 , and UGT1A1 genotypes

Variable	Overall (n = 93)				Carvedilol (n = 74)				Metoprolol (n = 19)
	Estimate	SE	p Value	Overall Genotype p Value	Estimate	SE	p Value	Overall Genotype p Value	Estimate	SE	p Value	Overall Genotype p Value
Model 1
Intercept	68.34	6.99	<0.0001		50.06	3.30	<0.0001		143.75	19.73	<0.0001
B1AR carrier	−0.08	9.34	0.99		2.68	4.44	0.55		−17.61	25.93	0.51
Model 2
Intercept	68.34	6.78	<0.0001		50.06	3.18	<0.0001		143.75	19.88	<0.0001
B1AR				0.044				0.036				0.56
Arg/Arg	Wild type				Wild type				Wild type
Arg/Gly	−6.12	9.38	0.52		0.27	4.38	0.95		−26.56	28.12	0.36
Gly/Gly	38.80	17.77	0.03		24.94	9.67	0.012		6.25	38.07	0.87
Model 3									NA ⁎
Intercept	53.24	10.45	<0.0001		50.31	4.61	<0.0001
CYP2D6				0.047				0.12
Extensive metabolizer	Wild type				Wild type
Poor metabolizer	46.76	18.47	0.013		19.69	9.45	0.04
Intermediate metabolizer	22.29	13.25	0.096		−4.36	6.34	0.49
Intermediate-to-extensive metabolizer	−4.13	15.55	0.79		−1.21	6.75	0.86
Ultrarapid metabolizer	17.73	13.44	0.19		4.21	6.12	0.49
Model 4
Intercept	65.32	4.73	<0.0001		50.21	2.21	<0.0001		130.47	14.02	<0.0001
CYP2D6—poor metabolizer	3	34.68	16.14	0.034	19.79	8.49	0.023		19.53	35.27	0.59
Model 5
Intercept	74.33	7.05	<0.0001		52.76	3.41	<0.0001		148.61	18.72	<0.0001
UGT1A1				0.44				0.55				0.55
Extensive metabolizer	Wild type				Wild type				Wild type
Poor metabolizer	−2.11	16.44	0.90		4.38	7.95	0.58		−23.61	43.90	0.60
Intermediate metabolizer	−12.31	9.74	0.21		−3.35	4.65	0.47		−29.86	27.29	0.29

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Relation of Body Mass Index to Mortality Among Men With Coronary Heart Disease
2. Comparison of Atorvastatin 80 mg/day Versus Simvastatin 20 to 40 mg/day on Frequency of Cardiovascular Events Late (Five Years) After Acute Myocardial Infarction (from the Incremental Decrease in End Points Through Aggressive Lipid Lowering [IDEAL] Trial)
3. All Red Yeast Rice Products Are Not Created Equal—or Legal
4. Usefulness of the Right Ventricular Systolic to Diastolic Duration Ratio to Predict Functional Capacity and Survival in Children With Pulmonary Arterial Hypertension

Stay updated, free articles. Join our Telegram channel

Join