Clinica Chimica Acta 445 (2015) 73–78
Contents lists available at ScienceDirect
Clinica Chimica Acta journal homepage: www.elsevier.com/locate/clinchim
Analysis of BAG3 plasma concentrations in patients with acutely decompensated heart failure☆,☆☆,★ Parul U. Gandhi a, Hanna K. Gaggin a, Arianna M. Belcher a, Jamie E. Harisiades a, Anna Basile b,c, Antonia Falco b,c, Alessandra Rosati b,c, Federico Piscione d, James L. Januzzi Jr. a,1, M. Caterina Turco c,d,⁎,1 a
Division of Cardiology, Massachusetts General Hospital, Boston, MA, United States Department of Pharmacy, University of Salerno, Fisciano, SA, Italy BIOUNIVERSA s.r.l., University of Salerno, Fisciano, SA, Italy d Department of Medicine and Surgery, University of Salerno, Baronissi, SA, Italy b c
a r t i c l e
i n f o
Article history: Received 26 November 2014 Received in revised form 24 February 2015 Accepted 25 February 2015 Available online 6 March 2015 Keywords: BAG3 BCL-2 athanogene 3 Heart failure Prognosis
a b s t r a c t Background: BCL-2-associated athanogene 3 (BAG3) is a protein implicated in the cardiomyocyte stress response and genesis of cardiomyopathy. Extracellular BAG3 is measurable in patients with heart failure (HF), but the relationship of BAG3 with HF prognosis is unclear. Methods: BAG3 plasma concentrations were measured in 39 acutely decompensated HF patients; the primary endpoint was death at 1 year. Baseline characteristics were compared by vital status and median BAG3 concentration. Correlation of BAG3 with left ventricular ejection fraction (LVEF) and other biomarkers was performed. Prognostic value was assessed using Cox proportional hazards regression and Kaplan–Meier analysis. Results: At baseline, median BAG3 was significantly higher in decedents (N = 11) than survivors (N = 28; 1489 ng/mL versus 50 ng/mL; P = 0.04); decedents also had worse renal function and higher median natriuretic peptide (NP) and sST2. BAG3 was not significantly correlated with NPs, mid-regional pro-adrenomedullin, sST2, or eGFR, however. Mortality was increased in patients with supra-median BAG3 (N 336 ng/mL; 42.1% versus 15.0%, P = 0.06). In age and LVEF-adjusted Cox proportional hazards, BAG3 remained a significant mortality predictor (HR = 3.20; 95% CI = 1.34–7.65; P = 0.02); those with supra-median BAG3 had significantly shorter time-to-death (P = 0.04). Conclusion: The stress response protein BAG3 is measurable in patients with ADHF and may be prognostic for death. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Abbreviations: HF, heart failure; BAG3, BCL-2-associated athanogene 3; CapZβ1, β1 subunit of F-actin capping protein; NYHA, New York Heart Association; BIONICS-HF, Biomonitoring and Cardiorenal Syndrome in Heart Failure; ADHF, acutely decompensated heart failure; EDTA, ethylenediaminotetraacetic acid; SD, standard deviation; IQR, interquartile range; LVEF, left ventricular ejection fraction; NT-proBNP, amino-terminal pro-Btype natriuretic peptide; BNP, B-type natriuretic peptide; MR-proANP, mid-regional proatrial natriuretic peptide; MR-proADM, mid-regional pro-adrenomedullin; sST2, soluble ST2; eGFR, estimated glomerular filtration rate; HFrEF, HF with reduced ejection fraction; HFpEF, HF with preserved ejection fraction. ☆ Funding sources: Dr. Gandhi is supported by the Dennis and Marilyn Barry Fellowship in Cardiology, Dr. Gaggin is supported in part by the Clark Fund for Cardiac Research Innovation, and Dr. Januzzi is supported in part by the Roman W. DeSanctis Clinical Scholar Endowment and the Hutter Family Professorship. ☆☆ Disclosures: Dr. Januzzi has received grant support from Thermo-Fisher, Singulex, and Siemens, and has received consulting income from Roche Diagnostics, Critical Diagnostics, and Sphingotec. Dr. Turco has received grant support from BIOUNIVERSA s.r.l. Dr. Gaggin has received consulting income from Roche Diagnostics, American Regent and Critical Diagnostics. ★ Clinical Trial Registration: Clinicaltrials.gov; NCT01570153. ⁎ Corresponding author at: University of Salerno, Via ponte don Melillo, 84084 Fisciano, Italy. Tel.: +39 089 969774; fax: +39 089 969602. E-mail address: [email protected] (M.C. Turco). 1 Contributed equally.
http://dx.doi.org/10.1016/j.cca.2015.02.048 0009-8981/© 2015 Elsevier B.V. All rights reserved.
As the number of patients with heart failure (HF) continues to grow, techniques to help improve our understanding of its pathophysiology may lead to superior strategies for risk prediction and therapy [1]. In this regard, study of biomarkers in HF has vastly expanded, with several new biomarkers under investigation for their use in diagnosis, prognosis, and management of HF [2,3]. Notably, most novel biomarkers in HF suffer from a relative lack of cardiovascular specificity, which may reduce their potential value. Accordingly, identification and assessment of biomarkers with greater cardiovascular specificity would be of considerable benefit. The BCL-2-associated athanogene (BAG) family consists of proteins that are involved in the regulation of heat shock proteins Hsp70 and HSc70 and possess a conserved C-terminus; members of this family differ based on the upstream region of the protein [4]. BAG3 is a member of this family; this protein is natively expressed in high concentrations in cardiac and skeletal muscle cells [5], and it has been implicated in apoptosis, cell adhesion, cytoskeleton remodeling, and autophagy [6]. BAG3 has also been linked to development of dilated cardiomyopathy [7]. Functionally, BAG3 may play a role in maintaining myofibril structural
74
P.U. Gandhi et al. / Clinica Chimica Acta 445 (2015) 73–78
integrity and Z-disc stability by its association with HSc70 and the β1 subunit of F-actin capping protein (CapZβ1) [8]. Recently, an extracellular form of BAG3 was found to be released from stressed cardiomyocytes in patients with chronic HF; furthermore, antibodies directed at BAG3 were also identified in these patients, implying a potential immune response [9]. Another report identified higher BAG3 concentrations in twenty patients with New York Heart Association (NYHA) class IV symptoms when compared to healthy controls or less symptomatic patients with HF, and elevated concentrations of BAG3 were associated with impending death or need for advanced HF therapies [10]. Given these promising results, further study is needed regarding measurement of BAG3 in HF. We therefore conducted a post hoc analysis of the Biomonitoring and Cardiorenal Syndrome in Heart Failure (BIONICS-HF) trial (Clinicaltrials.gov; NCT01570153), examining BAG3 measurement in a subgroup of the enrolled subjects [11]. 2. Materials and methods 2.1. Study procedures and patient population All procedures were approved by the Partners Healthcare Institutional Review Board, and informed consent was obtained from all patients. The details of the BIONICS-HF study have been reported previously [11]. In brief, this was originally a two center study involving the Massachusetts General Hospital and Ospedale Sant'Andrea whose focus was on identifying biomonitoring tools to better predict risk for HF complications, including renal failure and death. Patients were enrolled between April and July 2012 and were included if they presented with moderate to severe symptoms of acutely decompensated HF (ADHF) requiring intensification of diuretic therapy and were excluded if they required renal replacement therapy, received diuretic N 8 h prior to enrollment, or were unwilling or unable to participate in the study procedures. The primary endpoint of the BIONICS-HF study was worsening renal function and the secondary endpoint was a composite of worsening renal function and in-hospital mortality. With regard to the current analysis, data was collected at one year to evaluate the primary endpoint of death. At the time of presentation, a blood sample was collected for biomarker measurements. For the purposes of this analysis, thirty-nine subjects from the Massachusetts General Hospital cohort of the BIONICS-HF study with available blood samples for analysis of BAG3 were considered. 2.2. Sample analysis Blood was collected in ethylenediaminotetraacetic acid (EDTA) tubes and processed immediately with centrifugation at 4000 g for 15 min; samples were then aliquoted into tubes and stored at −80 °C prior to testing. A 10 mL sample of urine was also collected, spun for 10 min, and aliquoted into tubes for storage at −80 °C. BAG3 was measured using an enzyme linked immunosorbent assay (ELISA; BIOUNIVERSA s.r.l., Salerno, Italy). For the measurement of BAG3 protein content in plasma, we used NUNC Maxisorp 96 well ELISA plates coated with AC-rb1 antibody (BIOUNIVERSA s.r.l., Salerno, Italy) 4 μg/mL in PBS, pH. Plates were washed two times with washing buffer (PBS + 0.05% Tween-20), and then blocked for 1 h at room temperature with 0.5% fish gelatin in PBS (150 μL/well). Following blocking, the plates were washed two times with washing buffer, and plasma was diluted 1:50 with 0.5% fish gelatin in washing buffer and then applied (50 μL/well) in triplicates. After a 2 hour incubation at room temperature, the plates were washed six times. The biotinylated anti-BAG3 monoclonal antibody AC-rb3 (BIOUNIVERSA s.r.l., Salerno, Italy), 6 μg/mL, was added (50 μL/well); after a 1 hour incubation at room temperature, the plates were washed six times and HRP-conjugated avidin (eBioscience, San Diego, CA USA) diluted 1:500 was added (50 μL/well). Following a 30 minute incubation at room temperature, plates were washed six times. Subsequently, the reaction was developed with
TMB (50 μL/well) (eBioscience, San Diego, CA USA), stopped with 4.5 M sulfuric acid (25 μL/well) and the plates were analyzed spectrophotometrically at 450 nm. We used human recombinant (r) BAG3 protein at different concentrations (2 ng/mL to 128 ng/mL) as the standard curve; triplicate readings for each standard and plasma sample were averaged and subtracted from sample free of analytes (zero standard). A standard curve was constructed by plotting the mean adsorbance on the y axis against the concentration on the x axis. The obtained line equation was used to calculate BAG3 contents in wells and subsequently multiplied for the dilution factor. BAG3 plasma levels were expressed in ng/mL and the sensitivity limit was 100 ng/mL. rBAG3 protein concentration was obtained by measuring the adsorbance at 280 nm, while the purity was verified by SDS-PAGE (10%, reducing) and Coomassie stain comparing rBAG3 content to purified BSA. Samples with a BAG3 concentration at 4000 ng/mL have an interassay CV of b15%, determined according to Clinical and Laboratory Standards Institute (CLSI) guideline EP5-A2 [12]. Interassay CV calculations were also performed at BAG3 concentrations of 6442 ng/mL, 4086 ng/mL, and 320 ng/mL with CV of 8.8%, 13.8%, and 19.8%, respectively. The measuring range of the assay is 100 ng/mL to 25 μg/mL. In this analysis, any value below the lower limit of detection was modeled as 50% of the lower detection limit (thus, 50 ng/mL), consistent with multiple prior analyses [13–15]. In analyses of sample stability, BAG3 concentrations were previously found to be unaffected by two freeze– thaw cycles (M Turco, personal communication). Data regarding the linearity of this assay are shown in the Supplemental index. Detailed information on the other plasma and urine biomarkers examined in this analysis has been previously described [11,16]. 2.3. Statistical analysis Baseline characteristics were compared by vital status at 1 year as well as by the median BAG3 concentration of 336 ng/mL utilizing either Student's t test or Mann–Whitney U test for continuous variables and Chi square test or Fisher's exact test for categorical variables, as appropriate. Normally distributed variables were portrayed as mean ± standard deviation (SD) and non-normally distributed variables were expressed as median with interquartile range (IQR). BAG3 concentrations were also compared by left ventricular ejection fraction (LVEF), after log transformation with bivariate correlation. Receiver operating characteristic (ROC) curves were constructed for evaluation of BAG3 for the prediction of the gold standard outcome of mortality; from the ROC curve, area under the curve (AUC) for BAG3 was compared versus other markers. Additionally, sensitivity, specificity, positive predictive value and negative predictive value (PPV and NPV) were all estimated. Linear association between BAG3 and other continuous markers was examined using Pearson correlations; all biomarker values were similarly log transformed to achieve normality. Markers of interest included amino-terminal pro-B-type natriuretic peptide (NT-proBNP), B-type natriuretic peptide (BNP), mid-regional pro-atrial natriuretic peptide (MR-proANP) and mid-regional pro-adrenomedullin (MR-proADM), as well as soluble ST2 and estimated glomerular filtration rate (eGFR; calculated using the CKD-EPI equation) [17]. Cox-proportional hazards regression analysis was employed to examine the relationship between BAG3 and survival; for Cox modeling, non-normal variables were log-transformed, with hazard ratio (HR) expressed per standard deviation change. Initial models examined LnBAG3 alone, followed by models with age and LVEF as covariates. Finally, log-transformed concentrations of BNP, NT-proBNP, and sST2 were individually entered into the age and LVEF adjusted models. Kaplan– Meier methodology was used to investigate time to death for patients with BAG3 concentrations above and below the median, and significance was obtained using the log-rank test. All analyses were performed with SPSS Statistics Version 22.0 (Chicago, IL) with P values two-sided and significant if b0.05.
P.U. Gandhi et al. / Clinica Chimica Acta 445 (2015) 73–78
75
Table 1 Baseline characteristics divided by vital status at 1 year. Characteristic
Alive at 1 year N = 28
Deceased at 1 year N = 11
P value
Age (years), mean ± SD Male sex, N (%) Caucasian, N (%) Baseline EF (%), median (IQR) Hypertension, N (%) CKD, N (%) Prior HF, N (%) Prior MI, N (%) CAD, N (%) COPD, N (%) Tobacco use, N (%) Current Former DM, N (%) Atrial arrhythmia, N (%) NYHA class, N (%) II III IV SBP (mm Hg), mean ± SD DBP (mm Hg), mean ± SD Heart rate (bpm), median (IQR) Body mass index (kg/m2), median (IQR) JVD, N (%) Murmur, N (%) Rales, N (%) Edema, N (%) Cool extremities, N (%) ACE inhibitor, N (%) ARB, N (%) Beta-blocker, N (%) MRA, N (%) Loop diuretic, N (%) Aspirin, N (%) Digoxin, N (%) BUN (mg/dL), median (IQR) Creatinine (mg/dL), median (IQR) eGFR (mL/min/1.73 m2), mean ± SD NT-proBNP (pg/mL), median (IQR) BNP (pg/mL), median (IQR) MR-proANP (pmol/L), median (IQR) MR-proADM (pmol/L), median (IQR) BAG3 (ng/mL), median (IQR) sST2 (ng/mL), median (IQR) NGAL (pg/mL), median (IQR) Urine cystatin C (ng/mL), median (IQR)
69.1 ± 11.5 21 (75.0) 23 (82.1) 63.0 (58.0–67.0) 26 (92.9) 13 (46.4) 21 (75.0) 9 (32.1) 14 (60.7) 8 (28.6)
78.5 ± 12.1 8 (72.7) 11 (100.0) 44.5 (25.3–58.8) 9 (81.8) 7 (63.6) 10 (90.9) 6 (54.5) 8 (72.7) 1 (9.1)
0.03 1.0 0.30 b0.01 0.56 0.33 0.40 0.28 0.71 0.40 0.71
18 (64.3) 2 (7.1) 14 (50.0) 16 (57.1)
9 (81.8) 0 (0) 2 (18.2) 7 (63.6)
11 (39.3) 10 (35.7) 7 (25.0) 155.7 ± 32.9 78.2 ± 17.7 80.5 (61.8–86.0) 29.4 (26.5–35.7) 11 (39.3) 8 (28.6) 20 (71.4) 24 (85.7) 2 (7.1) 7 (25.0) 5 (17.9) 25 (89.3) 2 (7.1) 21 (75.0) 23 (82.1) 1 (3.6) 24.5 (16.0–41.8) 1.11 (0.81–1.8) 64.2 ± 30.0 2169 (1375–4692) 318 (190–707) 334.5 (241.3–474.5) 1.65 (1.36–1.86) 50.0 (50.0–2038.0) 79.1 (63.4–114.0) 174 (118–240) 242.3 (115.0–678.2)
2 (18.2) 6 (54.5) 3 (27.3) 121.6 ± 21.9 65.1 ± 10.4 84.0 (72.0–87.0) 25.8 (22.4–30.8) 6 (54.5) 7 (63.6) 9 (81.8) 6 (54.5) 0 (0) 4 (36.4) 1 (9.1) 11 (100.0) 3 (27.3) 9 (81.8) 8 (72.7) 2 (18.2) 42.0 (24.0–64.0) 1.49 (1.1–2.0) 42.5 ± 19.4 12,037 (2580–18,285) 497 (217–2100) 485.5 (347.5–863.0) 2.81 (2.33–3.76) 1489.0 (236.0–7357.0) 132.6 (98.9–196.7) 358 (159–386) 218.5 (28.2–534.5)
0.09 1.0 0.44
b0.01 0.03 0.37 0.05 0.48 0.07 0.69 0.09 1.0 0.69 0.66 0.55 0.13 1.0 0.66 0.19 0.03 0.09 0.01 b0.01 0.14 0.04 b0.001 0.04 0.02 0.01 0.45
SD = standard deviation, EF = ejection fraction, CKD = chronic kidney disease, HF = heart failure, MI = myocardial infarction, CAD = coronary artery disease, COPD = chronic obstructive lung disease, DM = diabetes mellitus, NYHA = New York Heart Association, SBP = systolic blood pressure, DBP = diastolic blood pressure, bpm = beats per minute, JVD = jugular venous distension, ACE = angiotensin converting enzyme, ARB = angiotensin receptor blocker, MRA = mineralocorticoid receptor antagonist, BUN = blood urea nitrogen, eGFR = estimated glomerular filtration rate, NT-proBNP = amino-terminal pro-B-type natriuretic peptide, BNP = B-type natriuretic peptide, MR-proANP = mid-regional pro-atrial natriuretic peptide, MR-proADM = mid-regional pro-adrenomedullin, sST2 = soluble ST2, NGAL = neutrophil gelatinase associated lipocalin.
3. Results Thirty-nine subjects were included in this analysis. At one year, twenty-eight patients were alive and eleven had died. Baseline characteristics divided by vital status at 1 year are shown in Table 1. Patients who expired were significantly older, had lower LVEF, and were more likely to have lower blood pressure and body-mass index and significantly higher sST2 and NT-proBNP concentrations. Additionally, those who were deceased also had significantly worse renal function and evidence of renal injury based on significantly elevated neutrophil gelatinase associated lipocalin (NGAL), compared to those who were alive. The median BAG3 concentration was significantly higher in those who died (1489 ng/mL) compared to those who were alive at 1 year (50 ng/mL; P = 0.04). Supplemental Table 2 portrays baseline characteristics divided by the median concentration of BAG3 (336 ng/mL). Patients with elevated BAG3
concentrations tended to have a more deleterious profile, with prevalent risk factors for adverse outcome in HF; while none of the plasma biomarkers were statistically significantly higher in those with elevated BAG3 concentrations, there was a numerical tendency towards higher natriuretic peptide and sST2 concentrations. Urinary cystatin C was more likely to be lower in those with supra-median BAG3 concentrations. BAG3 concentrations were then examined as a function of LVEF, dividing patients on the basis of LVEF ≥ 50% compared to those with LVEF below 50%. Although not statistically significant, there was a trend towards lower BAG3 concentrations in patients with an LVEF ≥ 50% (P = 0.11); the frequency of subjects with BAG3 concentrations below the lower detection limit also was higher in those with HF and preserved EF (HFpEF). A non-significant inverse correlation between log transformed LVEF and log transformed BAG3 was also seen (R = −0.19; P = 0.26); statistical significance was not reached partially due to the number of patients who had undetectable BAG3 concentrations (N = 18).
76
P.U. Gandhi et al. / Clinica Chimica Acta 445 (2015) 73–78
Table 2 Correlation of BAG3 with other established and emerging biomarkers. All biomarkers were log-transformed for normality. BAG3 BAG3
R P N NT-proBNP R 0.227 P 0.18 N 36 BNP R 0.171 P 0.31 N 38 MR-proANP R 0.266 P 0.10 N 39 MR-proADM R 0.098 P 0.55 N 39 sST2 R 0.214 P 0.19 N 39 eGFR R -0.246 P 0.13 N 39
NT-pro BNP
BNP
MR-pro ANP
MR-pro ADM
sST2
eGFR
0.227 0.18 36
0.171 0.31 38 0.757
Contents lists available at ScienceDirect
Clinica Chimica Acta journal homepage: www.elsevier.com/locate/clinchim
Analysis of BAG3 plasma concentrations in patients with acutely decompensated heart failure☆,☆☆,★ Parul U. Gandhi a, Hanna K. Gaggin a, Arianna M. Belcher a, Jamie E. Harisiades a, Anna Basile b,c, Antonia Falco b,c, Alessandra Rosati b,c, Federico Piscione d, James L. Januzzi Jr. a,1, M. Caterina Turco c,d,⁎,1 a
Division of Cardiology, Massachusetts General Hospital, Boston, MA, United States Department of Pharmacy, University of Salerno, Fisciano, SA, Italy BIOUNIVERSA s.r.l., University of Salerno, Fisciano, SA, Italy d Department of Medicine and Surgery, University of Salerno, Baronissi, SA, Italy b c
a r t i c l e
i n f o
Article history: Received 26 November 2014 Received in revised form 24 February 2015 Accepted 25 February 2015 Available online 6 March 2015 Keywords: BAG3 BCL-2 athanogene 3 Heart failure Prognosis
a b s t r a c t Background: BCL-2-associated athanogene 3 (BAG3) is a protein implicated in the cardiomyocyte stress response and genesis of cardiomyopathy. Extracellular BAG3 is measurable in patients with heart failure (HF), but the relationship of BAG3 with HF prognosis is unclear. Methods: BAG3 plasma concentrations were measured in 39 acutely decompensated HF patients; the primary endpoint was death at 1 year. Baseline characteristics were compared by vital status and median BAG3 concentration. Correlation of BAG3 with left ventricular ejection fraction (LVEF) and other biomarkers was performed. Prognostic value was assessed using Cox proportional hazards regression and Kaplan–Meier analysis. Results: At baseline, median BAG3 was significantly higher in decedents (N = 11) than survivors (N = 28; 1489 ng/mL versus 50 ng/mL; P = 0.04); decedents also had worse renal function and higher median natriuretic peptide (NP) and sST2. BAG3 was not significantly correlated with NPs, mid-regional pro-adrenomedullin, sST2, or eGFR, however. Mortality was increased in patients with supra-median BAG3 (N 336 ng/mL; 42.1% versus 15.0%, P = 0.06). In age and LVEF-adjusted Cox proportional hazards, BAG3 remained a significant mortality predictor (HR = 3.20; 95% CI = 1.34–7.65; P = 0.02); those with supra-median BAG3 had significantly shorter time-to-death (P = 0.04). Conclusion: The stress response protein BAG3 is measurable in patients with ADHF and may be prognostic for death. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Abbreviations: HF, heart failure; BAG3, BCL-2-associated athanogene 3; CapZβ1, β1 subunit of F-actin capping protein; NYHA, New York Heart Association; BIONICS-HF, Biomonitoring and Cardiorenal Syndrome in Heart Failure; ADHF, acutely decompensated heart failure; EDTA, ethylenediaminotetraacetic acid; SD, standard deviation; IQR, interquartile range; LVEF, left ventricular ejection fraction; NT-proBNP, amino-terminal pro-Btype natriuretic peptide; BNP, B-type natriuretic peptide; MR-proANP, mid-regional proatrial natriuretic peptide; MR-proADM, mid-regional pro-adrenomedullin; sST2, soluble ST2; eGFR, estimated glomerular filtration rate; HFrEF, HF with reduced ejection fraction; HFpEF, HF with preserved ejection fraction. ☆ Funding sources: Dr. Gandhi is supported by the Dennis and Marilyn Barry Fellowship in Cardiology, Dr. Gaggin is supported in part by the Clark Fund for Cardiac Research Innovation, and Dr. Januzzi is supported in part by the Roman W. DeSanctis Clinical Scholar Endowment and the Hutter Family Professorship. ☆☆ Disclosures: Dr. Januzzi has received grant support from Thermo-Fisher, Singulex, and Siemens, and has received consulting income from Roche Diagnostics, Critical Diagnostics, and Sphingotec. Dr. Turco has received grant support from BIOUNIVERSA s.r.l. Dr. Gaggin has received consulting income from Roche Diagnostics, American Regent and Critical Diagnostics. ★ Clinical Trial Registration: Clinicaltrials.gov; NCT01570153. ⁎ Corresponding author at: University of Salerno, Via ponte don Melillo, 84084 Fisciano, Italy. Tel.: +39 089 969774; fax: +39 089 969602. E-mail address: [email protected] (M.C. Turco). 1 Contributed equally.
http://dx.doi.org/10.1016/j.cca.2015.02.048 0009-8981/© 2015 Elsevier B.V. All rights reserved.
As the number of patients with heart failure (HF) continues to grow, techniques to help improve our understanding of its pathophysiology may lead to superior strategies for risk prediction and therapy [1]. In this regard, study of biomarkers in HF has vastly expanded, with several new biomarkers under investigation for their use in diagnosis, prognosis, and management of HF [2,3]. Notably, most novel biomarkers in HF suffer from a relative lack of cardiovascular specificity, which may reduce their potential value. Accordingly, identification and assessment of biomarkers with greater cardiovascular specificity would be of considerable benefit. The BCL-2-associated athanogene (BAG) family consists of proteins that are involved in the regulation of heat shock proteins Hsp70 and HSc70 and possess a conserved C-terminus; members of this family differ based on the upstream region of the protein [4]. BAG3 is a member of this family; this protein is natively expressed in high concentrations in cardiac and skeletal muscle cells [5], and it has been implicated in apoptosis, cell adhesion, cytoskeleton remodeling, and autophagy [6]. BAG3 has also been linked to development of dilated cardiomyopathy [7]. Functionally, BAG3 may play a role in maintaining myofibril structural
74
P.U. Gandhi et al. / Clinica Chimica Acta 445 (2015) 73–78
integrity and Z-disc stability by its association with HSc70 and the β1 subunit of F-actin capping protein (CapZβ1) [8]. Recently, an extracellular form of BAG3 was found to be released from stressed cardiomyocytes in patients with chronic HF; furthermore, antibodies directed at BAG3 were also identified in these patients, implying a potential immune response [9]. Another report identified higher BAG3 concentrations in twenty patients with New York Heart Association (NYHA) class IV symptoms when compared to healthy controls or less symptomatic patients with HF, and elevated concentrations of BAG3 were associated with impending death or need for advanced HF therapies [10]. Given these promising results, further study is needed regarding measurement of BAG3 in HF. We therefore conducted a post hoc analysis of the Biomonitoring and Cardiorenal Syndrome in Heart Failure (BIONICS-HF) trial (Clinicaltrials.gov; NCT01570153), examining BAG3 measurement in a subgroup of the enrolled subjects [11]. 2. Materials and methods 2.1. Study procedures and patient population All procedures were approved by the Partners Healthcare Institutional Review Board, and informed consent was obtained from all patients. The details of the BIONICS-HF study have been reported previously [11]. In brief, this was originally a two center study involving the Massachusetts General Hospital and Ospedale Sant'Andrea whose focus was on identifying biomonitoring tools to better predict risk for HF complications, including renal failure and death. Patients were enrolled between April and July 2012 and were included if they presented with moderate to severe symptoms of acutely decompensated HF (ADHF) requiring intensification of diuretic therapy and were excluded if they required renal replacement therapy, received diuretic N 8 h prior to enrollment, or were unwilling or unable to participate in the study procedures. The primary endpoint of the BIONICS-HF study was worsening renal function and the secondary endpoint was a composite of worsening renal function and in-hospital mortality. With regard to the current analysis, data was collected at one year to evaluate the primary endpoint of death. At the time of presentation, a blood sample was collected for biomarker measurements. For the purposes of this analysis, thirty-nine subjects from the Massachusetts General Hospital cohort of the BIONICS-HF study with available blood samples for analysis of BAG3 were considered. 2.2. Sample analysis Blood was collected in ethylenediaminotetraacetic acid (EDTA) tubes and processed immediately with centrifugation at 4000 g for 15 min; samples were then aliquoted into tubes and stored at −80 °C prior to testing. A 10 mL sample of urine was also collected, spun for 10 min, and aliquoted into tubes for storage at −80 °C. BAG3 was measured using an enzyme linked immunosorbent assay (ELISA; BIOUNIVERSA s.r.l., Salerno, Italy). For the measurement of BAG3 protein content in plasma, we used NUNC Maxisorp 96 well ELISA plates coated with AC-rb1 antibody (BIOUNIVERSA s.r.l., Salerno, Italy) 4 μg/mL in PBS, pH. Plates were washed two times with washing buffer (PBS + 0.05% Tween-20), and then blocked for 1 h at room temperature with 0.5% fish gelatin in PBS (150 μL/well). Following blocking, the plates were washed two times with washing buffer, and plasma was diluted 1:50 with 0.5% fish gelatin in washing buffer and then applied (50 μL/well) in triplicates. After a 2 hour incubation at room temperature, the plates were washed six times. The biotinylated anti-BAG3 monoclonal antibody AC-rb3 (BIOUNIVERSA s.r.l., Salerno, Italy), 6 μg/mL, was added (50 μL/well); after a 1 hour incubation at room temperature, the plates were washed six times and HRP-conjugated avidin (eBioscience, San Diego, CA USA) diluted 1:500 was added (50 μL/well). Following a 30 minute incubation at room temperature, plates were washed six times. Subsequently, the reaction was developed with
TMB (50 μL/well) (eBioscience, San Diego, CA USA), stopped with 4.5 M sulfuric acid (25 μL/well) and the plates were analyzed spectrophotometrically at 450 nm. We used human recombinant (r) BAG3 protein at different concentrations (2 ng/mL to 128 ng/mL) as the standard curve; triplicate readings for each standard and plasma sample were averaged and subtracted from sample free of analytes (zero standard). A standard curve was constructed by plotting the mean adsorbance on the y axis against the concentration on the x axis. The obtained line equation was used to calculate BAG3 contents in wells and subsequently multiplied for the dilution factor. BAG3 plasma levels were expressed in ng/mL and the sensitivity limit was 100 ng/mL. rBAG3 protein concentration was obtained by measuring the adsorbance at 280 nm, while the purity was verified by SDS-PAGE (10%, reducing) and Coomassie stain comparing rBAG3 content to purified BSA. Samples with a BAG3 concentration at 4000 ng/mL have an interassay CV of b15%, determined according to Clinical and Laboratory Standards Institute (CLSI) guideline EP5-A2 [12]. Interassay CV calculations were also performed at BAG3 concentrations of 6442 ng/mL, 4086 ng/mL, and 320 ng/mL with CV of 8.8%, 13.8%, and 19.8%, respectively. The measuring range of the assay is 100 ng/mL to 25 μg/mL. In this analysis, any value below the lower limit of detection was modeled as 50% of the lower detection limit (thus, 50 ng/mL), consistent with multiple prior analyses [13–15]. In analyses of sample stability, BAG3 concentrations were previously found to be unaffected by two freeze– thaw cycles (M Turco, personal communication). Data regarding the linearity of this assay are shown in the Supplemental index. Detailed information on the other plasma and urine biomarkers examined in this analysis has been previously described [11,16]. 2.3. Statistical analysis Baseline characteristics were compared by vital status at 1 year as well as by the median BAG3 concentration of 336 ng/mL utilizing either Student's t test or Mann–Whitney U test for continuous variables and Chi square test or Fisher's exact test for categorical variables, as appropriate. Normally distributed variables were portrayed as mean ± standard deviation (SD) and non-normally distributed variables were expressed as median with interquartile range (IQR). BAG3 concentrations were also compared by left ventricular ejection fraction (LVEF), after log transformation with bivariate correlation. Receiver operating characteristic (ROC) curves were constructed for evaluation of BAG3 for the prediction of the gold standard outcome of mortality; from the ROC curve, area under the curve (AUC) for BAG3 was compared versus other markers. Additionally, sensitivity, specificity, positive predictive value and negative predictive value (PPV and NPV) were all estimated. Linear association between BAG3 and other continuous markers was examined using Pearson correlations; all biomarker values were similarly log transformed to achieve normality. Markers of interest included amino-terminal pro-B-type natriuretic peptide (NT-proBNP), B-type natriuretic peptide (BNP), mid-regional pro-atrial natriuretic peptide (MR-proANP) and mid-regional pro-adrenomedullin (MR-proADM), as well as soluble ST2 and estimated glomerular filtration rate (eGFR; calculated using the CKD-EPI equation) [17]. Cox-proportional hazards regression analysis was employed to examine the relationship between BAG3 and survival; for Cox modeling, non-normal variables were log-transformed, with hazard ratio (HR) expressed per standard deviation change. Initial models examined LnBAG3 alone, followed by models with age and LVEF as covariates. Finally, log-transformed concentrations of BNP, NT-proBNP, and sST2 were individually entered into the age and LVEF adjusted models. Kaplan– Meier methodology was used to investigate time to death for patients with BAG3 concentrations above and below the median, and significance was obtained using the log-rank test. All analyses were performed with SPSS Statistics Version 22.0 (Chicago, IL) with P values two-sided and significant if b0.05.
P.U. Gandhi et al. / Clinica Chimica Acta 445 (2015) 73–78
75
Table 1 Baseline characteristics divided by vital status at 1 year. Characteristic
Alive at 1 year N = 28
Deceased at 1 year N = 11
P value
Age (years), mean ± SD Male sex, N (%) Caucasian, N (%) Baseline EF (%), median (IQR) Hypertension, N (%) CKD, N (%) Prior HF, N (%) Prior MI, N (%) CAD, N (%) COPD, N (%) Tobacco use, N (%) Current Former DM, N (%) Atrial arrhythmia, N (%) NYHA class, N (%) II III IV SBP (mm Hg), mean ± SD DBP (mm Hg), mean ± SD Heart rate (bpm), median (IQR) Body mass index (kg/m2), median (IQR) JVD, N (%) Murmur, N (%) Rales, N (%) Edema, N (%) Cool extremities, N (%) ACE inhibitor, N (%) ARB, N (%) Beta-blocker, N (%) MRA, N (%) Loop diuretic, N (%) Aspirin, N (%) Digoxin, N (%) BUN (mg/dL), median (IQR) Creatinine (mg/dL), median (IQR) eGFR (mL/min/1.73 m2), mean ± SD NT-proBNP (pg/mL), median (IQR) BNP (pg/mL), median (IQR) MR-proANP (pmol/L), median (IQR) MR-proADM (pmol/L), median (IQR) BAG3 (ng/mL), median (IQR) sST2 (ng/mL), median (IQR) NGAL (pg/mL), median (IQR) Urine cystatin C (ng/mL), median (IQR)
69.1 ± 11.5 21 (75.0) 23 (82.1) 63.0 (58.0–67.0) 26 (92.9) 13 (46.4) 21 (75.0) 9 (32.1) 14 (60.7) 8 (28.6)
78.5 ± 12.1 8 (72.7) 11 (100.0) 44.5 (25.3–58.8) 9 (81.8) 7 (63.6) 10 (90.9) 6 (54.5) 8 (72.7) 1 (9.1)
0.03 1.0 0.30 b0.01 0.56 0.33 0.40 0.28 0.71 0.40 0.71
18 (64.3) 2 (7.1) 14 (50.0) 16 (57.1)
9 (81.8) 0 (0) 2 (18.2) 7 (63.6)
11 (39.3) 10 (35.7) 7 (25.0) 155.7 ± 32.9 78.2 ± 17.7 80.5 (61.8–86.0) 29.4 (26.5–35.7) 11 (39.3) 8 (28.6) 20 (71.4) 24 (85.7) 2 (7.1) 7 (25.0) 5 (17.9) 25 (89.3) 2 (7.1) 21 (75.0) 23 (82.1) 1 (3.6) 24.5 (16.0–41.8) 1.11 (0.81–1.8) 64.2 ± 30.0 2169 (1375–4692) 318 (190–707) 334.5 (241.3–474.5) 1.65 (1.36–1.86) 50.0 (50.0–2038.0) 79.1 (63.4–114.0) 174 (118–240) 242.3 (115.0–678.2)
2 (18.2) 6 (54.5) 3 (27.3) 121.6 ± 21.9 65.1 ± 10.4 84.0 (72.0–87.0) 25.8 (22.4–30.8) 6 (54.5) 7 (63.6) 9 (81.8) 6 (54.5) 0 (0) 4 (36.4) 1 (9.1) 11 (100.0) 3 (27.3) 9 (81.8) 8 (72.7) 2 (18.2) 42.0 (24.0–64.0) 1.49 (1.1–2.0) 42.5 ± 19.4 12,037 (2580–18,285) 497 (217–2100) 485.5 (347.5–863.0) 2.81 (2.33–3.76) 1489.0 (236.0–7357.0) 132.6 (98.9–196.7) 358 (159–386) 218.5 (28.2–534.5)
0.09 1.0 0.44
b0.01 0.03 0.37 0.05 0.48 0.07 0.69 0.09 1.0 0.69 0.66 0.55 0.13 1.0 0.66 0.19 0.03 0.09 0.01 b0.01 0.14 0.04 b0.001 0.04 0.02 0.01 0.45
SD = standard deviation, EF = ejection fraction, CKD = chronic kidney disease, HF = heart failure, MI = myocardial infarction, CAD = coronary artery disease, COPD = chronic obstructive lung disease, DM = diabetes mellitus, NYHA = New York Heart Association, SBP = systolic blood pressure, DBP = diastolic blood pressure, bpm = beats per minute, JVD = jugular venous distension, ACE = angiotensin converting enzyme, ARB = angiotensin receptor blocker, MRA = mineralocorticoid receptor antagonist, BUN = blood urea nitrogen, eGFR = estimated glomerular filtration rate, NT-proBNP = amino-terminal pro-B-type natriuretic peptide, BNP = B-type natriuretic peptide, MR-proANP = mid-regional pro-atrial natriuretic peptide, MR-proADM = mid-regional pro-adrenomedullin, sST2 = soluble ST2, NGAL = neutrophil gelatinase associated lipocalin.
3. Results Thirty-nine subjects were included in this analysis. At one year, twenty-eight patients were alive and eleven had died. Baseline characteristics divided by vital status at 1 year are shown in Table 1. Patients who expired were significantly older, had lower LVEF, and were more likely to have lower blood pressure and body-mass index and significantly higher sST2 and NT-proBNP concentrations. Additionally, those who were deceased also had significantly worse renal function and evidence of renal injury based on significantly elevated neutrophil gelatinase associated lipocalin (NGAL), compared to those who were alive. The median BAG3 concentration was significantly higher in those who died (1489 ng/mL) compared to those who were alive at 1 year (50 ng/mL; P = 0.04). Supplemental Table 2 portrays baseline characteristics divided by the median concentration of BAG3 (336 ng/mL). Patients with elevated BAG3
concentrations tended to have a more deleterious profile, with prevalent risk factors for adverse outcome in HF; while none of the plasma biomarkers were statistically significantly higher in those with elevated BAG3 concentrations, there was a numerical tendency towards higher natriuretic peptide and sST2 concentrations. Urinary cystatin C was more likely to be lower in those with supra-median BAG3 concentrations. BAG3 concentrations were then examined as a function of LVEF, dividing patients on the basis of LVEF ≥ 50% compared to those with LVEF below 50%. Although not statistically significant, there was a trend towards lower BAG3 concentrations in patients with an LVEF ≥ 50% (P = 0.11); the frequency of subjects with BAG3 concentrations below the lower detection limit also was higher in those with HF and preserved EF (HFpEF). A non-significant inverse correlation between log transformed LVEF and log transformed BAG3 was also seen (R = −0.19; P = 0.26); statistical significance was not reached partially due to the number of patients who had undetectable BAG3 concentrations (N = 18).
76
P.U. Gandhi et al. / Clinica Chimica Acta 445 (2015) 73–78
Table 2 Correlation of BAG3 with other established and emerging biomarkers. All biomarkers were log-transformed for normality. BAG3 BAG3
R P N NT-proBNP R 0.227 P 0.18 N 36 BNP R 0.171 P 0.31 N 38 MR-proANP R 0.266 P 0.10 N 39 MR-proADM R 0.098 P 0.55 N 39 sST2 R 0.214 P 0.19 N 39 eGFR R -0.246 P 0.13 N 39
NT-pro BNP
BNP
MR-pro ANP
MR-pro ADM
sST2
eGFR
0.227 0.18 36
0.171 0.31 38 0.757
