Abstract
Background:
In HFrEF two clinical trials, the Beta-Blocker Evaluation of Survival Trial (BEST) and HF-ACTION, have reported an effectiveness interaction between the ADRB1 Arg389Gly polymorphism and beta-blockers (BBs). HF-ACTION additionally reported a dose-related interaction of unclear origin. If confirmed and pharmacogenetically resolved these findings may have important implications for HFrEF precision therapy. We used uniform methodology to investigate BB dose-ADRB1 Arg389Gly polymorphism interaction with major clinical endpoints in BEST/bucindolol and HF-ACTION/other BB databases.
Methods:
This was a retrospective analysis of prospectively designed DNA substudies from BEST (n=1040) and HF-ACTION (n=957). Subjects were genotyped for ADRB1 Arg389Gly and ADRA2C Ins322–325Del. BB dose was defined as either No/Low-Dose or High-Dose, according to total daily dose of either bucindolol (BEST subjects) or other BB (HF-ACTION subjects) standardized to carvedilol equivalents. The main outcome of interest was all-cause mortality (ACM) and CV mortality/HF Hospitalization was a secondary outcome.
Results:
Subjects in each trial had less ACM with High- vs. No/Low-Dose BB if they had ADRB1 Arg389Arg (BEST: HR=0.40, P=0.002; HF-ACTION: HR=0.45, P=0.005), but not Arg389Gly genotype (both P>0.2). Among gene-dose groups, there was a differential favorable treatment effect of 46% for High-Dose bucindolol with ADRB1 Arg389Arg vs. Gly carrier genotype (HR 0.54, P=0.018), but not for No/Low-Dose bucindolol. In contrast, HF-ACTION Arg389Arg genotype subjects taking No/Low-Dose BB had greater ACM compared to 389Gly carriers (HR=1.83, p=0.015), while ACM did not vary by genotype among subjects taking High-Dose BB (HR=0.84, p=0.55).
Conclusions:
The enhanced HFrEF efficacy of bucindolol in the ADRB1 Arg389Arg vs. 389Gly carrier genotypes occurs at high dose. Other BB taken at low dose have reduced efficacy for Arg389Arg genotype subjects compared to 389Gly carriers, suggesting a greater relative treatment effect at high dose. These data support guideline recommendations to use high, clinical trial “target” doses of all BBs to treat HFrEF.
Keywords: Genetics, Heart Failure, Translational Studies, Functional Genomics, pharmacogenetics, pharmacogenetics receptor, adrenergic receptor, beta-blocker, heart failure
INTRODUCTION
Bucindolol, investigated in the Beta-Blocker Evaluation of Survival Trial (BEST) for patients with advanced (NYHA III/IV) chronic heart failure with reduced ejection fraction (HFrEF), has been associated with improved survival compared to placebo in a subgroup of subjects with beta-1 adrenergic receptor (ADRB1) genotype Arg389Arg.1–3 This finding sparked interest to explore a possible clinical application of beta-blocker (BB) pharmacogenetics.4 However, most subsequent studies with other BBs have not evaluated differences in clinical outcomes by ADRB1 polymorphisms.5–9 The singular exception is the Heart Failure: A Controlled Trial Investigating Outcomes of Exercise Training (HF-ACTION) trial, in which the Arg389Arg but not 389Gly genotype was associated with mortality reduction for higher dose BBs.10, 11
The first objective of the current study was to examine dose-related aspects of BB interactions with the ADRB1 Arg389Gly polymorphism, emphasizing uniform methodology applied to the BEST/bucindolol and HF-ACTION/other BB databases. Because bucindolol, unlike other BB, has sympatholytic properties known to be modulated by the alpha2C adrenergic receptor (ADRA2C) Ins322–325Del polymorphism,12 we also investigated the impact of this polymorphism in both trials. A schematic representation of ADRB1 Arg389Gly and ADRAC Ins322–325Del polymorphisms on the human cardiac sympathetic neuroeffector system is given in eFigure1.
METHODS
A subset of the data generated for this study will be available from the corresponding author upon reasonable request. Protocols for HF-ACTION and BEST were reviewed and approved by the appropriate institutional review board or ethics committee for each participating center and by the coordinating center’s institutional review board. All patients provided written voluntary informed consent.
The Methods section of this study including scheme for BB dose standardization (eTable 1) is available in the Online Data Supplement or if viewed as a PDF, at the end of the file.
RESULTS
The analyzed DNA substudy populations included 1,035 of the 2,708 BEST entire cohort and 957 of 2331 total HF-ACTION subjects, representing participants in each trial that had successful ADRB1 Arg389Gly genotyping (100% in BEST, 98.1% in HF-ACTION) and a full complement of recorded variables used in covariate analysis. In BEST, 570 (55.1%) DNA substudy participants received no/low-dose bucindolol including 525 on placebo (92.1%), and had baseline characteristics similar to those taking high-dose bucindolol (Table 1). In HF-ACTION, which was not placebo controlled, the No/Low-Dose BB group (n=471, 49.2%) included 416 subjects (88.3%) on low-dose BB; these participants were overall older with more ischemic HF, and more likely to be non-African American leading to fewer with ADRA2C 322–325Del carrier genotype. These patterns were observed within ADRB1 and ADRA2C variants, with the additional finding of less aldosterone antagonist use among HF-ACTION subjects with homozygosity for either major allele (BEST, eTable 2; HF-ACTION, eTable 3).
Table 1.
Baseline characteristics of BEST and HF-ACTION DNA substudy participants.
| BEST | HF-ACTION | |||||
|---|---|---|---|---|---|---|
| Characteristic | No/Low Dose* N=570 |
High Dose* N=470 |
P-value | No/Low Dose* N=471 |
High Dose* N=486 |
P-value |
| Age (years) | 60 ± 13 | 60 ± 12 | .75 | 61 ± 12 | 57 ± 13 | <.0001 |
| Female (%) | 22% | 19% | .27 | 30% | 29% | .75 |
| African-Americans (%) | 19% | 21% | .24 | 24% | 36% | <.0001 |
| Ischemic etiology (%) | 60% | 56% | .21 | 58% | 43% | <.0001 |
| Left ventricular ejection fraction (%) | 24 ± 7 | 24 ± 7 | .89 | 25 ± 7 | 26 ± 8 | .033 |
| NYHA class (%) | .67 | .30 | ||||
| II | − | − | 64% | 69% | ||
| III | 93% | 92% | 35% | 31% | ||
| IV | 7% | 8% | 0.4% | 0.6% | ||
| Heart rate (bpm) | 81 ± 13 | 82 ± 14 | .09 | 71 ± 12 | 71 ± 11 | .50 |
| Systolic blood pressure (mmHg) | 117 ± 18 | 119 ± 18 | .06 | 114 ± 18 | 115 ± 19 | .29 |
| Blood urea nitrogen (mg/dL) | 24.2 ± 14.9 | 23.0 ± 12.4 | .14 | 25 ± 15 | 23 ± 13 | .017 |
| Serum creatinine (mg/dL) | 1.2 ± 0.4 | 1.2 ± 0.4 | .93 | 1.3 ± 0.8 | 1.3 ± 0.6 | .29 |
| HF therapies (%) | ||||||
| Beta-blocker (other than bucindolol) | − | 0.002 | ||||
| atenolol | − | − | 5% | 1% | ||
| bisoprolol | − | − | 1% | 1% | ||
| carvedilol | − | − | 56% | 59% | ||
| metoprolol | − | − | 8% | 10% | ||
| metoprolol XL | − | − | 29% | 29% | ||
| ACEI/ARB | 93% | 93% | .96 | 72% | 79% | .006 |
| Aldosterone receptor antagonist |
4% | 2% | .11 | 40% | 48% | .01 |
| Loop diuretic | 92% | 92% | .65 | 76% | 77% | .85 |
| Digoxin | 91% | 89% | .21 | 47% | 47% | .95 |
| ICD | 3% | 3% | .56 | 44% | 38% | .06 |
| CRT (-P or -D) | 0 | 0 | - | 17% | 16% | .63 |
| Adrenergic receptor genotype (%) | ||||||
| ADRB1 Arg389Arg | 45% | 50% | .16 | 46% | 45% | .80 |
| ADRB1 Arg389Gly | 55% | 50% | 54% | 55% | ||
| ADRA2C Ins322–325Ins | 81% | 80% | .70 | 76% | 68% | .005 |
| ADRA2C Ins322–325Del carrier | 19% | 20% | 24% | 32% | ||
No/Low Dose BB group: In BEST, 525 taking placebo and 45 taking ≤ 25 mg/day bucindolol. In HF-ACTION, 55 not taking BB and 416 taking ≤ 25 mg/day carvedilol-equivalents. High Dose BB group is defined as patients taking carvedilol-equivalent > 25 mg/day in HF-ACTION and > 25 mg/day of bucindolol in BEST
Abbreviations: ICD (implantable cardioverter defibrillator); CRT (cardiac resynchronization therapy with biventricular pacemaker).
Due to the degree of HF and relatively long follow-up, both DNA substudies had a large number of ACM (HF-ACTION: 125, BEST: 186) or CVM/HFH (HF-ACTION: 270, BEST 405) events (Tables 2 and 3).
Table 2.
Risk for Clinical Events with High vs. No/Low Dose Bucindolol in BEST subjects, by genotype.
| ACM | CVM/HFH | |||||
|---|---|---|---|---|---|---|
| Subgroup | No/Low Dose* | High Dose* | Hazard Ratio (95% CI)† | No/Low Dose* | High Dose* | Hazard Ratio (95% CI)† |
| All patients in DNA Substudy | 118/567 (21%) |
68/468 (15%) |
.62 (.46, .84) P=.002 |
255/567 (45%) |
150/468 (32%) |
.59 (.48, .73) P<.0001 |
| ADRB1 | ||||||
| 389Gly carrier | 63/308 (20%) |
44/235 (19%) |
.81 (.55, 1.20) P=.30 |
140/308 (45%) |
85/235 (36%) |
.69 (.52, .90) P=.007 |
| Arg389Arg | 55/259 (21%) |
24/233 (10%) |
.40 (.24, .65) P=.0002 |
115/259 (44%) |
65/233 (28%) |
.49 (.36, .67) P<.0001 |
| ADRA2C | ||||||
| 322–325Del carrier | 23/111 (21%) |
14/95 (15%) |
.62 (.31, 1.23) P=.20 |
49/111 (44%) |
31/95 (33%) |
.65 (.41, 1.03) P=.07 |
| Ins322–325Ins | 95/456 (21%) |
54/373 (14%) |
.63 (.45, .89) P=.008 |
206/456 (45%) |
119/373 (32%) |
.59 (.47, .74) P<.0001 |
| ADRA2C x ADRB1 | ||||||
| {322–325Del carrier + 389Gly carrier} |
14/79 (18%) |
10/54 (19%) |
1.06 (.43, 2.59) P=.90 |
32/79 (41%) |
16/54 (30%) |
.58 (.30, 1.10) P=.09 |
| {322–325Del carrier + Arg389Arg} |
9/32 (28%) |
4/41 (10%) |
.29 (.08, 1.09) P=.07 |
17/32 (53%) |
15/41 (37%) |
.50 (.24, 1.05) P=.07 |
| {Ins322–325Ins + 389Gly carrier} |
49/229 (21%) |
34/181 (19%) |
.85 (.54, 1.32) P=.5 |
108/229 (47%) |
69/181 (38%) |
.73 (.53, .99) P=.04 |
| {Ins322–325Ins + Arg389Arg} |
46/227 (20%) |
20/192 (10%) |
.40 (.23, .69) P=0.001 |
98/227 (43%) |
50/192 (26%) |
.46 (.33, .66) P<.0001 |
BEST placebo patients and bucindolol patients with exposure ≤ 25 mg/day are assigned to No/Low Dose Group. Patients with exposure >25 mg/day are assigned to High Dose group.
Generated by COX model with following covariates: serum creatinine, gender, ejection fraction, NYHA class, ischemic HF etiology, wide QRS, history of AF/AFL, aldosterone antagonist and angiotensin converting enzyme inhibitor. Parameterized to quantify benefit of high dose relative to low dose.
Table 3.
Risk for Clinical Events with High vs. No/Low Dose BB in HF-ACTION subjects, by genotype.
| ACM | CVM/HFH | |||||
|---|---|---|---|---|---|---|
| Subgroup | No/Low Dose* | High Dose* | Hazard Ratio (95% CI)† | No/Low Dose* | High Dose* | Hazard Ratio (95% CI)† |
| All patients in DNA Substudy | 74/428 (17%) | 51/435 (12%) | 0.73 (.50, 1.05) P=.09 | 146/428 (34%) | 124/435 (29%) | .87 (.68, 1.11) P=.27 |
| ADRB1 | ||||||
| 389Gly carrier | 33/231 (14%) | 31/240 (13%) | 1.05 (.63, 1.74) P=.86 | 78/231 (34%) | 77/240 (32%) | 1.08 (.78, 1.50) P=.64 |
| Arg389Arg | 41/197 (21%) |
20/195 (10%) | .45 (.26, .78) P=.005 | 68/197 (35%) | 47/195 (24%) | .63 (.43, .92) P=.018 |
| ADRA2C | ||||||
| 322–325Del carrier | 20/103 (19%) | 18/146 (12%) | .72 (.36, 1.45) P=.36 | 47/103 (46%) | 49/146 (34%) | .75 (.49, 1.15) P =.19 |
| Ins322–325Ins | 52/325 (16%) |
35/298 (12%) | .73 (.47 , 1.13) P=.157 | 97/325 (30%) |
78/298 (26%) | .89 (.65, 1.20) P=.44 |
| ADRA2C x ADRB1 | ||||||
| 322–325Del carrier + 389Gly carrier |
11/67 (16%) |
14/87 (16%) |
1.00 (.42, 2.41) P=.99 | 30/67 (45%) | 34/87 (39%) | 1.04 (.60, 1.78) P=.90 |
| 322–325Del carrier + Arg389Arg |
7/34 (21%) |
4/52 (8%) |
.27 (.05, 1.31) P = .10 | 15/34 (44%) |
13/52 (25%) |
.34 (.14, .82) P=.016 |
| Ins322–325Ins + 389Gly carrier |
20/155 (13%) | 17/151 (11%) | .92 (.47, 1.78) P=.79 | 44/155 (28%) |
42/151 (28%) | 1.07 (.69, 1.66) P=.75 |
| Ins322–325Ins + Arg389Arg |
32/160 (20%) |
16/138 (12%) | .49 (.26, .92) P=.027 | 51/160 (32%) | 34/138 (25%) | .68 (.43 , 1.06) P=.091 |
HF-ACTION patients with exposure ≤ 25 mg carvedilol equivalents daily are assigned to No/Low Dose BB Group, and those with greater exposure are assigned to High Dose BB Group.
Generated by COX model with following covariates: serum creatinine, gender, ejection fraction, NYHA class, ischemic HF etiology, wide QRS, history of AF/AFL, aldosterone antagonist and angiotensin converting enzyme inhibitor. Parameterized to quantify benefit of high dose relative to low dose.
ADRB1 Arg389Gly polymorphism
BEST/Bucindolol
For patients with ADRB1 Arg389Arg genotype there was a 60% ACM reduction in the High- vs. No/Low-Dose bucindolol groups (HR = 0.40, 95% CI 0.24, 0.65; P = 0.0002), but ADRB1 389Gly carriers did not have a similar benefit (P = 0.30; gene-dose interaction P = 0.035) (Table 2, Figure 1). Further, ADRB1 genotype predicted therapeutic effect only for subjects taking high-dose bucindolol (High-Dose Arg389Arg vs. 389Gly carriers HR = 0.54 (0.33, 0.90), P = 0.018), with the No/Low-Dose subgroups having similar ACM risk regardless of genotype (HR = 1.06 (0.73, 1.53), P = 0.77) (Figure 1). Low-dose (>0–25mg) Arg389Arg patients had a higher event rate than placebo patients (P = 0.006, eTable 4).
Figure 1.
All-cause mortality (AC0040044) in the BEST DNA Substudy (N=1035) by ADRB1 Arg389Gly polymorphism genotype and bucindolol dose. Varying dose within ADRB1 Arg389Arg (Arg/Arg) genotype: High Dose vs. No/ Low Dose, HR = 0.40 (95% CI 0.24, 0.65), P < 0.0001; ADRB1 389Gly carrier (Gly Carrier): High Dose vs. No/ Low Dose: HR = 0.81 (95% CI 0.55, 1.20), P = 0.30. Varying genotype within Dose group: Arg/ Arg vs. Gly Carrier in No/ Low Dose: HR = 1.06 (95% CI 0.73, 1.53), P =0.77; Arg/Arg vs. Gly Carrier in High Dose: HR = 0.54 (95% CI 0.33, 0.90), P = 0.018; Interaction P = .035.
CVM/ HFH event rates have not been previously reported for the BEST cohort, and HR are given in eTable 5 for Cox modeling using the trial’s statistical analysis plan2 and the more extensive covariate adjustment used in the current analysis. For both CVM/HFH and ACM, the two types of covariate adjustment yielded similar results.
Dose-genotype interactions for CVM/HFH (Table 2; eFigure 2) were consistent with observations for ACM, with the exception that ADRB1 389Gly carriers also had benefit with High- vs. Low/No-Dose bucindolol (HR = 0.69 (0.52, 0.90) P = 0.007).
HF-ACTION/Other BB
HF-ACTION subjects with Arg389Arg genotype also had a large reduction in ACM with high-dose BB compared to no/low-dose (HR 0.45 (0.26, 0.78), P = 0.005), in contrast to 389Gly carrier patients (HR = 1.05 (63, 1.74), P = 0.86; dose-genotype interaction P=0.02) (Table 3, Figure 2). However, in contrast to BEST findings, genotype differentiated risk for those taking no/low-dose BB, and not high-dose BB: Arg389Arg subjects on no/low-dose BB were at heightened ACM risk compared to Gly carriers (HR = 1.83 (1.13, 2.97), P = 0.015) (Figure 2; eTable 6). In the High-Dose BB group there was no evidence of a favorable pharmacogenetic effect on ACM in Arg389Arg vs. Gly carrier genotypes (HR 0.84 (0.47, 1.49), P = 0.55) (Table 3; Figure 2). The genotype-dose relationship was also examined by dose intervals. The Arg389Arg patients on no BB had a higher ACM event rate than the cognate 389Gly carrier group (34.8% vs. 12.5%, P = 0.048), although the comparison was based on only 12 events (eTable 6).
Figure 2.
All-Cause Mortality (ACM) in the HF-ACTION DNA Substudy (N = 863) by ADRB1 Arg389Gly polymorphism genotype and beta-blocker (BB) dose (carvedilol equivalents). Varying dose within ADRB1 Arg389Arg (Arg/Arg) genotype: High Dose vs. No/ Low Dose BB, HR = 0.45 (95% CI 0.26, 0.78), P = 0.005; ADRB1 389Gly carrier (Gly Carrier): High Dose vs No/ Low Dose BB, HR = 1.05 (95% CI 0.63, 1.74), P = 0.86. Varying genotype within Dose group: Arg/ Arg vs. Gly Carrier in No/ Low Dose BB: HR = 1.83 (95% CI 1.13, 2.97), P = 0.015; Arg/ Arg vs. Gly Carrier in High Dose BB: HR = 0.84 (95% CI 0.47, 1.49), P = 0.55; Interaction P = 0.02.
For CVM/HFH, within ADRB1 genotypes the effects of High- vs. No/Low-Dose were similar to those for ACM (Table 3, eFigure 3). The Arg389Arg vs. 389Gly carrier comparison within High- and No/Low-Dose groups was also similar to ACM data, except in the No/Low-Dose group an increased CVM/HFH rate in Arg389Arg vs. 389Gly carrier genotypes was not observed (HR = 1.28 with 95% CI of .91–1.80, P = 0.16) (eFigure 3).
ADRA2C Ins322–325Del polymorphism
BEST/Bucindolol
For ACM there was no effect of high- vs. no/low-dose bucindolol when subjects were grouped by ADRA2C genotype (HR = 0.62 for Del carriers, HR = 0.63 for Ins/Ins, HR = 0.59 for entire DNA substudy cohort) or CVM/HFH (HR = 0.65 for Del carriers, HR = 0.59 for Ins/Ins, HR = 0.87 for entire DNA substudy cohort) (Table 2).
HF-ACTION/Other BB
Similar to BEST/bucindolol, high-dose BB did not affect ACM in subjects when grouped by ADRA2C genotype (HR = 0.72 for Del carriers, HR = 0.59 for Ins322–325Ins, HR = 0.73 for entire DNA substudy cohort) or CVM/HFH (HR = 0.75 for Del carriers, HR = 0.89 for Ins322–325Ins, HR = 0.87 for entire DNA substudy cohort) (Table 3).
Combinatorial Genotypes
BEST/Bucindolol
When assessed by ADRA2C genotype combinations, ADRB1 Arg389Arg but not 389Gly carrier combinations showed major dose-relatedness. A complete loss of efficacy for high-dose bucindolol was observed for the {ADRA2C Del carrier + ADRB1 Gly carrier} genotype (HR 1.06 (0.43, 2.59), P = 0.90) (Table 2).
HF-ACTION/Other BB
For ACM and CVM/HFH, there was no evidence that ADRA2C Ins322–325Del genotype influenced any of the ADRB1 Arg389 genotype/dose interactions (Table 3).
DISCUSSION
For treatment with bucindolol and other BBs, ADRB1 genotypes and BB dose significantly interacted with ACM risk. In both trials, high-dose BB was much more effective than no/low dose in the ADRB1 Arg389Arg genotype, but not in the ADRB1 389Gly carriers. The importance of BB dose may explain the lack of association between ADRB1 genotype and clinical outcomes among all BB-treated patients in prior studies. However, the differential genotype-dose effect on overall survival was qualitatively different in the two trials. In the High vs. No/Low-dose bucindolol comparison, the differential favorable effect of ADRB1 Arg389Arg (HR = 0.40, P <.0001) compared to 389Gly carrier (HR = 0.81, P = 0.30) for ACM was due to an Arg389Arg vs. 389Gly effect in high-dose patients (reduction in ACM by 46%). In contrast, the HF-ACTION genotype-dose interaction for high- vs. no/low-dose BB (ADRB1 Arg389Arg HR = 0.45, P = 0.005, compared to 389Gly carrier HR = 1.05, P = 0.86) was not due to differences in Arg389Arg vs. 389Gly genotype in high-dose patients (HR 0.84, P = 0.55), but rather to an increase in ACM in the No/Low-Dose group (Arg389Arg vs. 389Gly carriers HR 1.83, P = 0.015). Thus HFrEF patients do not receive differential effects by ADRB1 genotype with high-dose BB. Rather, among patients taking low-dose BB, there is evidence of increased ACM if ADRB1 genotype is Arg389Arg vs. 389Gly. Results for the secondary endpoint of CVM/HFH supported these findings.
In addition to ADRB1 Arg389Gly genotypes, we investigated BB dose effects related to the ADRA2C Ins322–325Del polymorphism, where the 322–325Del allele is associated with dysregulation of norepinephrine release.12, 13 For bucindolol vs. placebo, ADRA2C Ins322–325Ins genotype irrespective of dose has been shown to identify enhanced vs. no clinical response12 and differentiate non-response from intermediate response among ADRB1 Arg389Gly carrier patients.2 In the current analysis of high- vs. no/low dose bucindolol, the ADRA2C Ins322–325Del polymorphism did not differentiate response to bucindolol, either alone or in combination with ADRB1 389 genotype. In HF-ACTION, ADRA2C 322–325 genotype did not affect clinical outcomes response to other BB, in agreement with a previous study.14 Although a prior epidemiological study demonstrated a synergistic relationship between ADRB1 Arg389Arg and ADRA2C Del322–325Del for development of HFrEF,15 our findings do not support use of ADRA2C genotype to predict optimal BB dose for clinical outcomes.
Effects of high-dose bucindolol have been previously described for HF endpoints and arrhythmias, and are attributed to selective inverse agonism and/or norepinephrine-lowering properties impacting high norepinephrine affinity ADRB1 Arg389 receptors.2, 3, 16, 17 However, the finding of an increase in ACM in ADRB1 Arg389Arg vs. 389Gly carrier subjects receiving low-dose other BBs has not been reported, and is of unclear origin. There was also a hint of a similar effect in ADRB1 Arg389Arg BEST patients receiving >0–25 mg low-dose bucindolol, but the limited number of ACM events (N=10) precluded interpretation. The mechanistic possibilities for a differential increase in ACM in ADRB1 Arg389Arg patients receiving low-dose BBs include: 1) selection of patients in the No/Low-dose group who could not tolerate high-dose BBs, where due to differences in HF severity the ADRB1 Arg389Arg genotype was associated with worse outcomes compared to 389Gly carriers, and 2) an adverse pharmacologic effect of low-dose other BBs in ADRB1 Arg389Arg patients. Regarding the former possibility, covariate adjustment was performed for most factors that might influence outcome, but Cox modeling may not have not adequately corrected for increased risk in the Low/No-Dose group. Alternatively, up-regulation of desensitized beta-1 adrenergic receptor signal transduction is a possibility for an adverse pharmacological effect of No/Low-Dose other BB,18 when combined with inadequate receptor occupancy for sufficient beta-blockade. Regardless of the mechanism, for ACM we have demonstrated that unlike for bucindolol in BEST, the previously reported10 pharmacogenetically greater therapeutic effect of other BBs with high-dose BB in HFrEF patients with an ADRB1 Arg389Arg genotype is not due to favorable effects vs. 389Gly carriers, but rather to worse outcomes in the Arg389Arg No/Low-Dose group.
Limitations
The data sources were from clinical trial substudies, but the number of ACM or CVM/HFH events was substantial and in the range of what is reported in entire cohorts of some Phase 3 trials.19, 20 The two analyzed DNA substudies, both prospectively designed, represent the largest two cohorts of HFrEF patients with genotyping for ADRB1 polymorphisms, documentation of BB use, and assessment of standard clinical HFrEF endpoints. Although the methodologic approach was the same between the two databases, since BEST was a placebo-controlled trial and HF-ACTION was not there were more patients on no BB in BEST vs. HF-ACTION. However, effects of low-dose bucindolol in ADRB1 Arg389Arg vs. Arg389Gly genotypes were similar to that for placebo (enhanced efficacy in the Arg389Arg genotype was observed only for individual dose subgroups >25mg). Although reasons for not being on a BB differ by trial, we examined genotype differences within dosing interval in each trial separately. Finally, the combinatorial genotype analysis was limited by small sample sizes for several subgroups.
Conclusions
Beta-blocker dose effects on ACM or CVM/HFH risk may be observed for HFrEF patients with an ADRB1 Arg389Arg genotype. The observed decreased risk compared to the alternative 389Gly carrier genotype may be due a true reduction in event rates at high/target doses (bucindolol in BEST) or an increase in event rates with lower doses (other BB in HF-ACTION). These data support guideline recommendations that in HFrEF beta-blockers should be used at the higher, target doses used in all positive Phase 3 trials. The potential role for ADRB1 genotype to assess risk with low-dose BB needs prospective study.
Supplementary Material
Clinical Implications.
The ADRB1 Arg389Arg vs. its counterpart Gly carrier genotype is most useful to predict enhanced efficacy of bucindolol taken at high doses. In contrast, HFrEF subjects with the Arg389Arg genotype taking usual care, other BBs at low dose have significantly worse survival than their 389Gly counterparts. These observations highlight a potential role for pharmacogenomics in clinical practice to predict optimal BB dose.
Acknowledgments
Acknowledgements: Author Contributions: KSP conceived of the study design, and wrote all manuscript drafts. MRB edited all manuscript drafts and had full access to all the data in the study and takes responsibility for the integrity of the data plus the accuracy of the data analysis. GD performed the Cox modeling of the data as well as all statistical analyses, and edited the manuscript. SBL was the Principal Investigator (PI) and WTA the Co-PI of the BEST Adrenergic Receptor Polymorphism DNA substudy, and both reviewed and edited the manuscript. MN performed initial statistical analyses for the different outcomes assessed in the HF-ACTION DNA substudy and reviewed the manuscript. GMF and KA led the collection of DNA samples for the HF-ACTION DNA substudy and reviewed the manuscript. CMO was the PI of the HF-ACTION trial, and CMO and MF contributed to study design and data analysis and reviewed and edited the manuscript. Disclaimer: The views expressed in this manuscript are those of the authors and do not necessarily represent the views of the National Heart, Lung, and Blood Institute, or the National Institutes of Health. Role of the Sponsors: The funding agencies were not directly involved in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; or the decision to submit the manuscript for publication.
Sources of Funding: The study was funded by grant 5T32GM086330–05 from the National Institute of General Medical Sciences, and ARCA biopharma, Inc. The HF-ACTION biobank was funded by the Duke Clinical Research Institute and HF-ACTION biobank study subinvestigators. The HF-ACTION trial was funded by grants HL063747 and HL093374 from the National Heart, Lung and Blood Institute. BEST was funded by the Veteran’s Administration Cooperative Studies Program, the National Heart, Lung and Blood Institute, Incara Pharmaceutical Co, and ARCA biopharma, Inc.
Footnotes
Disclosures: Mona Fiuzat is a consultant for FDA; Gordon Davis and Michael Bristow are employees of ARCA biopharma, the sponsor of bucindolol; Michael Bristow and Stephen Liggett are founders of ARCA biopharma; and William Abraham and Stephen Liggett are consultants to ARCA biopharma.
REFERENCES:
- 1. Beta-Blocker Evaluation of Survival Trial I. A trial of the beta-blocker bucindolol in patients with advanced chronic heart failure. N Engl J Med. 2001;344:1659–1667. [DOI] [PubMed] [Google Scholar]
- 2. O’Connor CM, et al. Combinatorial pharmacogenetic interactions of bucindolol and beta1, alpha2c adrenergic receptor polymorphisms. PLoS One. 2012;7:e44324. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Liggett SB, et al. A polymorphism within a conserved beta(1)-adrenergic receptor motif alters cardiac function and beta-blocker response in human heart failure. Proc Natl Acad Sci U S A. 2006;103:11288–11293. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Parikh KS, et al. Potential applications of pharmacogenomics to heart failure therapies. Heart Fail Clin. 2014;10:599–606. [DOI] [PubMed] [Google Scholar]
- 5. Liu WN, et al. Beta1 adrenergic receptor polymorphisms and heart failure: A meta-analysis on susceptibility, response to beta-blocker therapy and prognosis. PLoS One. 2012;7:e37659. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Metra M, et al. Role of beta-adrenergic receptor gene polymorphisms in the long-term effects of beta-blockade with carvedilol in patients with chronic heart failure. Cardiovasc Drugs Ther. 2010;24:49–60. [DOI] [PubMed] [Google Scholar]
- 7. Savva J, et al. Polymorphisms of adrenoceptors are not associated with an increased risk of adverse event in heart failure: A merit-hf substudy. J Card Fail. 2009;15:435–441. [DOI] [PubMed] [Google Scholar]
- 8. Terra SG, et al. Beta-adrenergic receptor polymorphisms and responses during titration of metoprolol controlled release/extended release in heart failure. Clin Pharmacol Ther. 2005;77:127–137. [DOI] [PubMed] [Google Scholar]
- 9. White HL, et al. An evaluation of the beta-1 adrenergic receptor arg389gly polymorphism in individuals with heart failure: A merit-hf sub-study. Eur J Heart Fail. 2003;5:463–468. [DOI] [PubMed] [Google Scholar]
- 10. Fiuzat M, et al. Association between adrenergic receptor genotypes and beta-blocker dose in heart failure patients: Analysis from the hf-action DNA substudy. Eur J Heart Fail. 2013;15:258–266. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. O’Connor CM, et al. Efficacy and safety of exercise training in patients with chronic heart failure: Hf-action randomized controlled trial. JAMA. 2009;301:1439–1450. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Bristow MR, et al. An alpha2c-adrenergic receptor polymorphism alters the norepinephrine-lowering effects and therapeutic response of the beta-blocker bucindolol in chronic heart failure. Circ Heart Fail. 2010;3:21–28. [DOI] [PubMed] [Google Scholar]
- 13. Hein L, et al. Two functionally distinct alpha2-adrenergic receptors regulate sympathetic neurotransmission. Nature. 1999;402:181–184. [DOI] [PubMed] [Google Scholar]
- 14. Sehnert AJ, et al. Lack of association between adrenergic receptor genotypes and survival in heart failure patients treated with carvedilol or metoprolol. J Am Coll Cardiol. 2008;52:644–651. [DOI] [PubMed] [Google Scholar]
- 15. Small KM, et al. Synergistic polymorphisms of beta1- and alpha2c-adrenergic receptors and the risk of congestive heart failure. N Engl J Med. 2002;347:1135–1142. [DOI] [PubMed] [Google Scholar]
- 16. Aleong RG, et al. Adrenergic receptor polymorphisms and prevention of ventricular arrhythmias with bucindolol in patients with chronic heart failure. Circ Arrhythm Electrophysiol. 2013;6:137–143. [DOI] [PubMed] [Google Scholar]
- 17. Aleong RG, et al. Prevention of atrial fibrillation by bucindolol is dependent on the beta(1)389 arg/gly adrenergic receptor polymorphism. JACC Heart Fail. 2013;1:338–344. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Gilbert EM, et al. Comparative hemodynamic, left ventricular functional, and antiadrenergic effects of chronic treatment with metoprolol versus carvedilol in the failing heart. Circulation. 1996;94:2817–2825. [DOI] [PubMed] [Google Scholar]
- 19. Gheorghiade M, et al. Effect of aliskiren on postdischarge mortality and heart failure readmissions among patients hospitalized for heart failure: The astronaut randomized trial. JAMA. 2013;309:1125–1135. [DOI] [PubMed] [Google Scholar]
- 20. Zannad F, et al. Eplerenone in patients with systolic heart failure and mild symptoms. N Engl J Med. 2011;364:11–21. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.