An α2C-Adrenergic Receptor Polymorphism Alters the Norepinephrine-Lowering Effects and Therapeutic Response of the β-Blocker Bucindolol in Chronic Heart FailureCLINICAL PERSPECTIVE
Background— Adrenergic activation is an important determinant of outcomes in chronic heart failure. Adrenergic activity is regulated in part by prejunctional α2C-adrenergic receptors (ARs), which exhibit genetic variation in humans. Bucindolol is a novel β-AR blocking agent that also lowers systemic norepinephrine and thus is also a sympatholytic agent. This study investigated whether α2C-AR polymorphisms affect sympatholytic effects of bucindolol in patients with heart failure.
Methods and Results— In the β-Blocker Evaluation of Survival Trial, adrenergic activation was estimated by systemic venous norepinephrine measured at baseline, 3 months, and 12 months posttreatment in patients treated with placebo or bucindolol. In the β-Blocker Evaluation of Survival Trial AR polymorphism substudy, DNA was collected from 1040 of the 2708 randomized patients, and α2C-AR gene polymorphisms (α2C Del322-325 or the wild-type counterpart) were measured by polymerase chain reaction and gel electrophoresis. Patients who were α2C Del carriers (heterozygotes or homozygotes) exhibited a much greater sympatholytic response to bucindolol (decrease in norepinephrine at 3 months of 153±57 pg/mL, P=0.012 compared with placebo versus decrease of 50±13 pg/mL in α2C wild type, P=0.0005 versus placebo; P=0.010 by interaction test). α2C Del carriers had no evidence of a favorable survival benefit from bucindolol (mortality compared with placebo hazard ratio, 1.09; 95% CI, 0.57 to 2.08; P=0.80), whereas bucindolol-treated subjects who were wild type for the α2C-AR had a 30% reduction in mortality (hazard ratio, 0.70; 95% CI, 0.51 to 0.96; P=0.025).
Conclusions— In the β-Blocker Evaluation of Survival Trial AR polymorphism substudy, the norepinephrine lowering and clinical therapeutic responses to bucindolol were strongly influenced by α2C receptor genotype.
Received June 9, 2009; accepted October 22, 2009.
Adrenergic activity is a major determinant of outcome in chronic heart failure (CHF).1,2 Multiple lines of evidence indicate that sustained high levels of β-adrenergic activity produce adverse effects on myocardial structure and function.3,4 These observations form the basis of the rationale for β-blocker therapy of heart failure.4 On the other hand, a certain amount of adrenergic activity is required to support the failing heart, and too rapid or extreme removal of adrenergic drive can result in adverse outcomes.5–7
Clinical Perspective on p 21
The regulation of cardiac adrenergic activity is complex, involving mechanisms modulating central sympathetic outflow, norepinephrine (NE) neuronal synthesis, prejunctional NE release, and neuronal reuptake of NE.3 Adrenergic activity is likely also influenced by genetic variation, particularly in adrenergic receptors (ARs) that regulate NE release. One such receptor is the α2C-AR, which is present on prejunctional adrenergic nerve terminals where it provides tonic inhibition of NE release.8 Combined genetic ablation of the α2C- and α2A-ARs in mice results in increased plasma NE levels and, ultimately, a cardiomyopathy.8 Mice with isolated α2C gene ablation also exhibit adrenergic dysfunction, with alterations in NE release in response to nerve stimulation9 and pharmacological modulation of evoked NE release.8,10 In addition, α2C-AR knockout mice have increased circulating levels10 and adrenal release11 of epinephrine. In humans, functionally important polymorphic variation in the α2C-AR exists in the form of a 4-amino acid deletion variant (α2c Del322-325, corresponding to nucleotides +964 to 975)9,12 that exhibits marked loss of function when expressed in model systems.9 The prevalence of this genetic variant is enriched in African American (AA) populations, where it has an allele frequency of ≈0.40 compared with ≈0.04 in whites.9,12 Based on the observation that NE release by isolated atria is increased in α2c-AR knockout mice,8,10 it has been predicted12 that the α2C Del322-325 variant in humans would be associated with increased systemic NE levels, particularly in situations such as heart failure where central sympathetic outflow is increased. Indeed, there is evidence that the α2C Del322-325 polymorphism leads to an increased adrenergic activity in normal subjects.13 However, there is conflicting experimental evidence for the α2C Del322-325 variant being associated with an increased adrenergic activity in patients with heart failure.14,15
We have previously reported that patients with advanced CHF who have marked NE lowering in response to treatment with the β-blocker/sympatholytic agent bucindolol exhibit increased mortality and heart failure hospitalizations compared with patients with little or no sympatholytic response.7 In contrast, patients who had a mild decrease in NE have an enhancement of therapeutic effects of bucindolol.7 These sympatholytic responses to bucindolol have the potential to be influenced by polymorphisms of prejunctional, NE release regulating ARs, including the α2C Wt/Del polymorphism.
For these reasons, we hypothesized that genetic variation in α2c-ARs could have influenced the clinical and/or sympatholytic response to bucindolol in β-Blocker Evaluation of Survival Trial (BEST), including racial heterogeneity of therapeutic effects.16 To test this general hypothesis, we measured α2C genotypes in participants in the BEST DNA substudy17 and assessed their relationships with systemic venous NE levels at 3 months and subsequent clinical responses to bucindolol.
Clinical Protocol Subject Population
The BEST protocol18 and the main results as determined by an unadjusted statistical analysis16 have been previously described. A summary of these results as well as results obtained using covariate adjustment and cardiac transplant censoring is given in supplemental Table I.
NE Sample Collection and Measurements
Peripheral venous NE samples were drawn at baseline, 3, and 12 months and measured by an high-performance liquid chromatography technique, as previously described.7 In this report, only baseline and changes at 3 months are reported. In the α2c-AR substudy reported here, 875 of 1040 patients had baseline NE levels drawn, and 740 patients had NE levels drawn at both baseline and after 3 months of study medication.
DNA Collection and Extraction
The BEST DNA substudy consisted of 1040 patients enrolled in the clinical trial who also signed a written consent for DNA collection as previously described.17
AR Polymorphism Substudy of BEST
DNA was obtained from the BEST DNA Bank by submitting a grant application for a competitive review by the BEST DNA Oversight Committee. The approved application contained a hypothesis based on the preliminary data, a sample size calculation, and a statistical analysis plan.
Measurement of α2C 322-325 AR Gene Variants
Genotyping of the α2c-AR gene was performed based on the differential size of polymerase chain reaction products for the wild type and the 12 nucleotide deletion polymorphism, as described in the Data Supplement.
Affinity and Agonist Activity of Bucindolol for α2C-AR Wild Type and Del322-325 Polymorphisms
The affinity (Ki) of bucindolol for α2C 322-325 AR variants was determined by transfecting the cDNAs of each cDNA into Chinese hamster ovary cells as previously described,9 with further details given in the Data Supplement.
The primary end points used in this study were prespecified in the BEST DNA Bank AR Polymorphism grant application that was approved by the BEST Trial DNA Oversight Committee and are time to all-cause mortality (the primary end point in BEST),19 time to all-cause mortality or cardiac transplantation, and left ventricular ejection fraction (LVEF) change measured at 12 months. Secondary end points were time to cardiovascular mortality (as adjudicated by the BEST End Points Committee), time to heart failure progression (with the prespecified definition of death due to heart failure, cardiac transplantation, hospitalization due to heart failure, or a hospital emergency department visit for treatment of heart failure not requiring hospitalization), the number of days each patient was hospitalized for the treatment of heart failure, and systemic plasma venous NE levels.
In addition to α2C, 5 other AR polymorphisms were included in the AR polymorphism substudy, and a prespecified α level of 0.01 (2 tailed) was used to indicate a statistical significance. P values of ≤0.05 and >0.025 were considered as indicating a tendency toward significance, and P values ≤0.025 and >0.010 were considered a strong tendency toward significance. Because the BEST DNA substudy began after the main protocol, the L-truncation statistical method19,20 is additionally presented as an alternative method of analysis to control for sampling bias. Further details of the statistical methods are given in the Data Supplement.
Allele Frequency and Study Population Characteristics
The frequency of the α2C Del322-325 allele was 0.12 in all patients, 0.04 in non-AAs, and 0.43 in AAs (P<0.0001 versus non-AA). The baseline demographic and other characteristics data in the 1040 subjects in the DNA substudy are given in supplemental Table II. There were no important clinical differences between the 2 treatment groups. Importantly, there were also no clinically relevant differences between the 1040 patients in the DNA substudy and the 2708 patients in the entire cohort.16,17 Supplemental Table III gives the baseline characteristic or demographic data by α2C genotype. As can be observed, the only substantial difference between the major allele homozygote (α2C Wt/Wt) and the α2C Del carrier (heterozygotes or Del/Del homozygotes) group is that the Del carrier group comprise a higher percentage of AA patients and other factors (hypertension, nonischemic etiology, younger age, history of hypertension, or higher blood pressure) that track with heart failure in AA patients.
Effect of α2C-AR Polymorphisms on Baseline NE Levels
As shown in Table 1, baseline NE mean values determined in 875 patients in the DNA substudy were 475±276 pg/mL in the placebo group (n=436) and 500±289 pg/mL in the bucindolol group (n=439, P=0.088 versus placebo for Ln-transformed data). There was no difference in baseline NE in α2CDel322-325 carriers (520±351 pg/mL) versus α2C Wt/Wt patients (480±265 pg/mL), P=0.69. Because of the lack of gene-dose effects and the relatively small number of Del322-325 homozygotes, Del322-325 homozygotes and heterozygotes were combined in a “Del322-315 carrier” subgroup for further analyses.
Alpha2c Del322-325 Genotype Is Associated With Large Decreases in NE in the Bucindolol Group
Table 2 presents the change in NE at 3 months by α2c-AR genotype and race. For the entire DNA substudy sample, in placebo-treated patients, there was a small (19.5 pg/mL), nonsignificant (P=0.19) increase in NE as would be expected in a population of patients with advanced CHF. In contrast, in patients receiving bucindolol, there was a reduction in NE of 68.7 pg/mL at 3 months (P=0.0001 versus baseline, P<0.0001 versus placebo change). In placebo-treated patients, at 3 months, the Del carrier group had a nonsignificantly (P=0.44) larger increase (by 41.0 pg/mL) in NE as compared with the α2C Wt/Wt group. In bucindolol-treated patients a comparison of the α2C Wt/Wt and α2C Del 322-325 carrier genotypes (Table 2) revealed that Del carriers had a 3.1-fold greater reduction in NE compared with the α2c wild-type subgroup (P=0.001). If the changes in the respective placebo groups are considered, α2C Del carriers have a 3.3-fold greater decrease in NE at 3 months as compared with α2C Wt/Wt bucindolol-treated patients. The test for interaction between the 2 treatment groups and genotypes yielded a P=0.010. Thus, the differential lowering of NE by bucindolol in patients with α2C Del 322-325 carriers versus Wt/Wt genotypes meets the prespecified level of statistical significance for a treatment × genotype interaction.
As can be observed in Table 2, both non-AA and AA patients had placebo-subtracted reductions in NE in the bucindolol group that were nominally greater in the α2C Del322-325 carriers than in the α2C Wt/Wt genotype (non-AA, Del322-325 carriers=−168 pg/mL, α2C Wt/Wt=−63.3 pg/mL, P=0.19 by interaction test; AA, Del322-325 carriers=−221 pg/mL, α2C Wt/Wt=63.9 pg/mL, P=0.24 by interaction).
In the covariate-adjusted model, the placebo-subtracted lowering of NE at 3 months in α2C Wt/Wt patients was by 63 pg/mL in the bucindolol group (P=0.002 versus placebo) and in α2C Del carrier patients by 182 pg/mL (P=0.014 versus placebo), with a treatment group × genotype group interaction P value of 0.017.
Effect of α2c-AR Variants on Heart Failure Natural History
Kaplan-Meier curves for the end points of all-cause mortality (Figure A), cardiovascular mortality (Figure B), the combined end points of all-cause mortality or cardiac transplantation (Figure C), and heart failure progression (Figure D) plotted by α2C-AR genotype in placebo-treated patients indicate that there is no statistically significant difference between genotypes (respective hazard ratios [HRs; wild type/Del322-325] 1.22, 1.23, 1.27, and 1.11; P values ranging from 0.34 to 0.54).
Mortality and Morbidity Benefits of Bucindolol by α2C-Receptor Genotype
The Figure also presents Kaplan-Meier curves for patients treated with bucindolol, by α2c-AR genotype subgroups. HRs comparing bucindolol with placebo and number of events are given in Table 3 in the “Primary Model” columns. For the all-cause mortality end point, in wild-type homozygotes, bucindolol produced a strong tendency toward significance (P=0.025) for a reduction in mortality by 30% (Figure A and Table 3). In contrast, in Del322-325 carriers, there was a nonsignificant (P=0.79), 9% increase in mortality in the bucindolol group (Figure A and Table 3). For cardiovascular mortality (Figure B and Table 3), bucindolol was associated with 41% reduction in event rate (P=0.003) in α2C wild-type homozygotes but no evidence of benefit in α2C Del322-325 carriers (HR=1.03; P=0.93). For the combined end point of mortality or transplantation (Figure C and Table 3), in wild-type homozygotes, bucindolol treatment was associated with a 32% reduction (P=0.014) in event rate, whereas in Del322-325 carriers, there was no evidence of any bucindolol efficacy (HR=1.03, P=0.93). For the heart failure progression end point (Figure D, Table 3), in patients with the α2C Wt/Wt genotype, there was a 28% reduction (P=0.002) in event rate in the bucindolol group, whereas the α2C Del322-325 genotype group had only a small (13%) nonsignificant (P=0.52) tendency toward significance for a reduction in event rate. The number of days hospitalized for heart failure per patient (Table 3) was reduced by 35% in the α2C Wt/Wt bucindolol group and was not obviously affected (3.2% increase in the bucindolol versus placebo group) in patients carrying the α2C Del322-325 allele. A test for interaction between the 2 treatment groups and the 2 polymorphisms was not significant for any end point, with P values ranging from 0.15 (cardiovascular mortality) to 0.45 (heart failure progression). The evidence of a differential clinical efficacy effect by α2C polymorphism consists of achieving or having a strong tendency toward significance (P≤0.025) for the prespecified α of 0.01 in the α2C Wt/Wt genotype but not in α2C Del322-325 carrier group. This occurred for all 5 clinical end points, with cardiovascular mortality and heart failure progression achieving a P≤0.01. Although the smaller n (207 versus 833 in α2C Wt/Wt) in the α2C Del322-325 carrier group would preclude a statistical significance for anything other than HRs deviating by large margins from 1.0, the proximity of the HRs to 1.0 (0.87 to 1.09), and the minimal effect on hospitalization days per patient (3.2%) indicates that a P<0.01 would not have been achieved in the α2C Del carrier group even if the sample size had been in the same range as for α2C Wt/Wt.
Because the α2C deletion polymorphism is racially distributed, we performed a covariate adjustment for race, and for gender, etiology, and LVEF as well. The results were quite similar to the unadjusted analysis: all-cause mortality, α2C Wt/Wt bucindolol/placebo HR=0.73 (0.52 to 1.01), P=0.053 (152 events), α2C 322-325 Del carrier HR=0.96 (0.48 to 1.95), P=0.92 (37 events); cardiovascular mortality, α2C Wt/Wt HR=0.60 (0.42 to 0.0.87), P=0.006 (127 events); α2C 322-325 Del carrier HR=1.00 (0.48 to 2.09), P=1.00 (32 events); all-cause mortality or cardiac transplantation, α2C Wt/Wt HR=0.70 (0.51 to 0.96), P=0.025 (169 events), α2C 322-325 Del carrier HR=0.96 (0.48 to 1.95), P=0.92 (38 events); and heart failure progression, α2C Wt/Wt HR=0.72 (0.58 to 0.92), P=0.003 (353 events), α2C Del carrier HR=0.93 (0.59 to 1.47), P=0.76 (83 events). In addition, in the Data Supplement, an unadjusted analysis of the 5 clinical end points is performed by α2C genotypes, within racial subgroups. The results within the non-AA racial subgroup are consistent with the results for all races described in the Figure and Table 3. In the AA subgroup, the results between α2C genotypes were inconsistent, with no evidence of a differential response in mortality driven time to event end points, and some statistically insignificant numeric evidence of differential effects in the 2 heart failure hospitalization end points but in opposite directions. In AA α2C Del carriers compared with Wt homozygotes, there was a greater therapeutic effect of bucindolol in heart failure progression but a lesser effect in heart failure hospitalization days per patient. These inconsistent effects by genotype in the AA racial subgroup are likely due to small numbers of patients and events.
Analysis of Clinical End Points Using Left Truncation
Given in Table 3 are time-to-event analyses using the L-truncation model, which minimizes bias due to “late entry” of patients into the DNA substudy, due to the 2-year gap between the start of the BEST study, and the start of the BEST DNA substudy. For the heart failure hospitalization-driven end point of heart failure progression, this method has lower power because of the elimination of 72 nonmortality events from patients who had DNA drawn at study close-out visits or after cardiac transplantation. As can be seen in Table 3, L-truncation gives HR point estimates that are similar to the primary model. Further discussion of these results is given in the Data Supplement.
Relative Effect Sizes by Genotype
Table 4 gives the relative effect sizes (RES) values for the 5 clinical end points by α2C genotype. RES values >1.0 mean the effect size is greater in the genotype group than in the entire DNA substudy cohort, and negative values indicate that the bucindolol event rate was higher than in the placebo group. Average RES was calculated across the 5 clinical end points. Collectively, average RES provides a better measure of efficacy than each individual end point, although it should be noted that some of these end points are not strictly independent (eg, mortality is a component of 4 of the 5 end points, and HF hospitalization is a component of 2 of the 5). For patients with the α2C Wt/Wt genotype, the average RES across the 5 clinical end points was 1.25, meaning there was a 25% enhancement in efficacy. In contrast, in the 19.9% of the patients who were carrying a Del322-325 allele, there was a complete loss of efficacy, with an average RES of −0.02. This means that there was a 127% differential efficacy between α2C Wt/Wt and α2C Del carrier patients. The covariate-adjusted analysis results were similar, with a 110% differential efficacy. Both analysis models indicate that there is a mild enhancement of clinical efficacy in patients who are homozygous for the wild-type α2C receptor and a loss of efficacy in patients who are α2C Del carriers.
LVEF results by α2C genotype are given in the Data Supplement. At 12 months, there were no genotype-related differences in LVEF improvement in the bucindolol versus placebo groups.
Affinity of Bucindolol for α2c-AR Variants
The affinity of bucindolol for α2c-AR variants was measured in vitro, as described in the Data Supplement. Bucindolol had low, micromolar affinity for both wild-type and Del322-325 α2c-ARs.
The data presented in this study support the following conclusions: (1) A 4-amino acid deletion (Del322-325) in the α2C-AR, a polymorphism enriched in AAs, does not obviously influence baseline adrenergic activity in patients with CHF who have elevated adrenergic drive as the result of a heart failure compensatory mechanism; (2) the α2C Del322-325 polymorphism is associated with a marked increases in the sympatholytic response to bucindolol; (3) there is no impact of α2C 322-325 polymorphisms on heart failure natural history in an established population with CHF, at least over a mean follow-up of 24 months in patients treated with placebo; and (4) the clinical response to bucindolol is related to the presence or absence of the α2c Del322-325 allele, which presumably influences clinical outcomes by regulating the sympatholytic response to bucindolol.
The major new finding in this study is that the sympatholytic effects of bucindolol are under genetic control by α2C 322-325 Wt/Del receptor variants. Within the bucindolol group, there was a 3.1-fold greater reduction in NE at 3 months in patients who were α2C Del carries compared with Wt/Wt patients (153±57 versus 50±13 pg/mL versus P=0.001). The placebo-adjusted NE decrease in bucindolol-treated α2C Del carriers was 3.3-fold greater than in α2C Wt/Wt patients, which was statistically significant (P=0.010) by test for interaction. These data indicate that in patients with wild-type α2C receptors and normal α2C regulation of neuronal NE release, bucindolol is only mildly sympatholytic. In contrast, in patients with heart failure who have α2C Del genotypes and receptors, bucindolol is a moderate sympatholytic agent, with its NE lowering properties presumably sensitized by loss of function in prejunctional α2C receptors.
In addition to being associated with a much greater sympatholytic response, α2C Del carrier patients treated with bucindolol had no evidence of any clinical efficacy. Using a method that references clinical responses to those in the entire DNA substudy, patients who were α2C Del carriers had a complete loss of efficacy across the 5 clinical end points measured. In contrast, patients with the α2C Wt/Wt genotype had a mild enhancement of clinical efficacy by 25%. The clinical results by α2C polymorphism were not substantively influenced by delays in the DNA drawn in the majority of randomized patients, as L-truncation, which minimizes bias in this setting,19,20 yielded mortality end point and heart failure hospitalization days per patient results that were quite similar to the primary method of analysis. These data are consistent with a previous analysis of the effects of NE reduction on bucindolol efficacy in BEST,3 where mild reductions of NE averaging 44 pg/mL produced effects on mortality that were superior to moderate-marked reductions (reductions of >244 pg/mL) or moderate increases in NE. The differential association of α2C polymorphisms with clinical effects was more impressive than effects on LVEF, where there were only small differences in LVEF change between the α2C Wt/Wt and α2C Del carrier groups.
A relationship of large reductions in NE and serious adverse events has been observed for higher doses of moxonidine in the MOXSE [MOXonidine Safety and Efficacy study] trial,21 and in the MOXCON [MOXonidine CONgestive Heart Failure trial],6 these same higher doses produced an increase in mortality and heart failure hospitalizations. In the BEST DNA substudy, the Del322-325 α2C-AR polymorphism was associated with a decrease in NE of 141 pg/mL compared with baseline and 182 pg/mL compared with placebo change at 3 months. The larger amount of NE lowering associated with the α2C Del322-325 allele likely led to an increase in heart failure clinical end point event rates, effectively canceling any beneficial effects of bucindolol mediated by β-blockade. On the basis of previous studies,6,7,21 there is obvious clinical/mechanistic plausibility for a differential clinical response between patients who have large versus small sympatholytic responses to drug therapy. The mechanistic basis for mild NE lowering adding to the therapeutic effects of β-blockade may be that in patients with heart failure receiving full-dose β-blockade, NE signaling is not completely abolished,22 and this leaves room for an additional therapeutic response to mild (<100 pg/mL) NE lowering.
The precise mechanism of the pharmacological interaction of the Del322-325 polymorphism with bucindolol was not determined in this study. However, a differential effect of bucindolol acting directly on the α2CAR receptor variants can be excluded, because bucindolol exhibited low (micromoles per liter) affinities for the wild-type and Del322-325 receptor variants that were within 2-fold of one another. It is clear that the Del322-325 polymorphism greatly impairs the signal transduction function of the α2C receptor,9 and prejunctional α2C AR dysfunction8,10 may predispose to greater sympatholytic effects from agents acting on β2-AR or other prejunctional mechanisms that regulate NE release. Bucindolol is the only β-blocking agent that substantially lowers systemic adrenergic drive.7,23 As a result, other β-blockers such as metoprolol and carvedilol would not be expected to exhibit the interaction of α2C wild-type or Del polymorphisms with clinical end points, which seems to be the case.24
Because the α2C Del322-325 allele is highly enriched in the AA population, it is important to account for potentially confounding effects of race when studying the α2C wild type or Del polymorphism. In this study, this was done by analyzing racial subgroups separately and by performing a covariate adjustment for race. The much greater amount of NE lowering by bucindolol in α2C Del322-325 patients was present in both non-AA (reduced by 2.65-fold>wild-type homozygotes) and AA (reduced by 3.46-fold>wild-type homozygotes). Covariate adjustment by race did not appreciably affect outcomes in any analyzed end point, and clinical outcome results stratified by race were indeterminate in AA patients because of small numbers of events. Thus, the end point modifying effects of the α2C Del322-325 allele seem to be independent of AA versus non-AA status, but this issue should be further investigated in adequately powered dedicated studies.
The α2C Wt/Del322-325 polymorphism is the second AR gene variant to be associated with a differential clinical response to bucindolol in patients with heart failure. The other AR polymorphism is the β1 389 Arg/Gly polymorphism,17 which interacts with the β-AR blocking effects of bucindolol rather than NE-releasing properties of bucindolol. This creates the potential for a clinical end point therapeutic interaction between the 2 sets of polymorphisms, which is the subject of active investigation. From the standpoint of both natural history and the therapeutic effects of β-blockade, the potential exists for the α2C Del322-325 polymorphism to intergenically interact with polymorphisms in β1-ARs,12,25 and this is the subject of ongoing investigation.
In summary, the data presented in this study support the hypothesis that α2c-AR genetic variation can modify the NE lowering and therapeutic responses to bucindolol and potentially other sympatholytic agents. Although the findings are based on the associations of AR polymorphic variants with changes in NE and clinical end points, the results are supported by both biological plausibility and data from other clinical studies. Finally, although the study was prospective and achieved a prespecified statistical result from the standpoint of NE changes, as for all studies replication in a second study would be ideal. However, such a study will not have the advantage of a placebo control inasmuch as β-blockade is now recommended treatment for heart failure, and hence, any confirmatory study will necessarily have a different design.
We thank Rachel Rosenberg for her assistance in editing and manuscript handling.
Sources of Funding
The study was funded by the Veteran’s Administration Cooperative Studies Program, the National Heart Lung and Blood Institute, National Institute Health, Incara Pharmaceutical Co, and ARCA biopharma, Inc.
Drs Bristow and Port are part-time employees of ARCA Biopharma, Inc; Drs Davis and Murphy are employees of ARCA; and Dr Bristow is a company director. They and Dr Liggett have equity in the company. Drs Lavori, Liggett, and Robertson are consultants to ARCA.
The online-only Data Supplement is available at http://circheartfailure.ahajournals.org/cgi/content/full/CIRCHEARTFAILURE.109.885962.
- ↵Bristow MR. β-adrenergic receptor blockade in chronic heart failure. Circulation. 2000; 101: 558–569.
- ↵Bristow MR, Krause-Steinrauf H, Nuzzo R, Liang Cheng-Seng, Lindenfeld J, Lowes BD, Hattler B, Abraham WT, Olson L, Krueger S, Thaneemit-Chen S, Hare JM, Domanski MJ, Eichhorn EJ, Lavori P, Zelis R. Effect of baseline or changes in adrenergic activity on clinical outcomes in the beta-blocker evaluation of survival trial (BEST). Circulation. 2004; 110: 1437–1442.
- ↵Small KM, Forbes SL, Rahman RF, Bridges KM, Liggett SB. A four amino acid deletion polymorphism in the third intracellular loop of the human α2c-adrenergic receptor confers impaired coupling to multiple effectors. J Biol Chem. 2000; 275: 23059–23064.
- ↵Brede M, Wiesmann F, Jahns R, Hadamek K, Arnolt C, Neubauer S, Lohse MJ, Hein L. Feedback inhibition of catecholamine release by two different alpha2-adrenoceptor subtypes prevents progression of heart failure. Circulation. 2002; 106: 2491–2496.
- ↵Gilsbach R, Brede M, Beetz N, Moura E, Muthig V, Gerstner C, Barreto F, Neubauer S, Vieira-Coelho MA, Hein L. Heterozygous alpha 2C-adrenoceptor-deficient mice develop heart failure after transverse aortic constriction. Cardiovasc Res. 2007; 75: 728–737.
- ↵Neumeister A, Charney DS, Belfer I, Geraci M, Holmes C, Sharabi Y, Alim T, Bonne O, Luckenbaugh DA, Manji H, Goldman D, Goldstein DS. Sympathoneural and adrenomedullary functional effects of alpha2C-adrenoreceptor gene polymorphism in healthy humans. Pharmacogenet Genomics. 2005; 15: 143–149.
- ↵Liggett SB, Mialet-Perez J, Thaneemit-Chen S, Weber SA, Greene SM, Hodne D, Nelson B, Morrison J, Domanski MJ, Abraham WT, Anderson JL, Carlquist JF, Krause-Steinrauf HJ, Lazzeroni LC, Port JD, Lavori PW, Bristow MR. A polymorphism within a highly conserved β1-adrenergic receptor motif alters β-blocker response in multiple models and human heart failure. Proc Natl Acad Sci USA. 2006; 103: 11288–11293.
- ↵Hougaard P. Analysis of Multivariate Survival Data. New York, NY: Springer; 2000: 72–73, 84–85.
- ↵Jamison RL, Shih MC, Humphries DE, Guarino PD, Kaufman JS, Goldfarb DS, Warren SR, Gaziano JM, Lavori P; Veterans Affairs Site Investigators. Effect of the MTHFR C677T and A1298C polymorphisms on survival in patients with advanced CKD and ESRD: a prospective study. Am J Kidney Dis. 2009; 53: 779–789.
- ↵Swedberg K, Bristow MR, Cohn JN, Dargie H, Wright TJ; for the MOXSE Investigators. Effects of sustained-release moxonidine, and imidazoline agonist, on plasma norepinephrine in patients with congestive heart failure. Circulation. 2002; 105: 1797–1803.
- ↵Lowes BD, Wolfel GW, Minobe WA, Lindenfeld JA, Robertson AD, Gilbert EM, Bristow MR. Evidence for ongoing β-adrenergic signaling of gene expression and heart rate in CHF subjects treated with β-blocking agents. Circulation. 2001; 104: II-436.
- ↵Sehnert AJ, Daniels SE, Elashoff M, Wingrove JA, Burrow CR, Horne B, Muhlestein JB, Donahue M, Liggett SB, Anderson JL, Kraus WE. Lack of association between beta adrenergic receptor genotype and survival in heart failure patients treated with carvedilol or metoprolol. J Am Coll Cardiol. 2008; 52: 644–651.
Drugs that have been shown to be effective in a disease population work in only a subset, typically estimated at 30% to 50% of the total. It has long been a dream of those involved in drug development to be able to more accurately define a drug’s responsive subpopulation, ideally by a widely available and accurate biomarker. The α2C-adrenergic receptor regulates norepinephrine (NE) release from adrenergic nerve terminals, including those in the heart. Bucindolol is a novel β-adrenergic receptor blocking agent that reduces NE release, the only β-blocking agent that does so. This reduction in NE by bucindolol is modulated by α2C-adrenergic receptors for which there is a relatively common (≈10% of the general population) minor polymorphic allele that leads to a 4 amino acid deletion in the receptor protein, resulting in nearly total loss of function. When this allele is present in either a single (heterozygotes) or double copy, dysregulation of NE release results. This dysregulation leads to a substantial increase in inhibition of NE release of bucindolol. As a result, NE levels precipitously fall to levels below that which can support cardiac function, predisposing the patient to adverse events that cancel the beneficial effects of β-blockade. On the other hand, when the major “wild type” allele is present in the homozygous state, bucindolol produces only mild and clinically beneficial degrees of NE lowering. Thus, the use of genetic testing could improve the clinical response to bucindolol, which is a strategy being pursued in the development of this drug.