Abstract
Objectives
The N-terminal fragment of pro-B-type natriuretic peptide (NT-proBNP) is a widely used heart failure (HF) biomarker. Commercial NT-proBNP immunoassays detect only a subfraction of endogenous NT-proBNP, as the antibodies target a region of NT-proBNP that could be glycosylated at Ser44. The diagnostic utility of immunoassays measuring total NT-proBNP remains unclear.
Methods
NT-proBNP was measured in 183 HF and 200 non-HF patients diagnosed by two independent cardiologists blinded to NT-proBNP results. Plasma samples either non-treated or treated with a mixture of glycosidases were analyzed by the Elecsys proBNP II assay (Roche Diagnostics, based on antibodies targeting a glycosylated region of NT-proBNP) and the SuperFlex NT-proBNP assay (PerkinElmer, based on antibodies targeting regions of NT-proBNP that are free of O-glycans). The diagnostic accuracy of the two assays was analyzed by comparison of ROC curves.
Results
The ROC-AUC for the proBNP II assay was 0.943 (95% CI 0.922–0.964) for NT-proBNP measured in untreated samples and 0.935 (0.913–0.958) for NT-proBNP measured in glycosidase-treated samples. The SuperFlex NT-proBNP assay in untreated samples gave a ROC-AUC of 0.930 (95% CI 0.907–0.954). The median percentage of non-glycosylated NT-proBNP to total NT-proBNP was 1.5–1.6-fold lower in the non-HF group compared to that in the HF group.
Conclusions
The clinical value of total NT-proBNP for HF diagnosis was similar to the subfraction of NT-proBNP that was non-glycosylated at Ser44. The lower percentage of non-glycosylated NT-proBNP to total NT-proBNP in non-HF patients suggests that total NT-proBNP might be more sensitive in individuals without current or prior symptoms of HF.
Introduction
The B-type natriuretic peptide (BNP) along with N-terminal fragment of pro-B-type natriuretic peptide (NT-proBNP) are accepted as clinically useful biomarkers for heart failure (HF) diagnosis, management, and risk assessment for ruling out HF in the emergency department or in the outpatient setting [1], [2], [3].
All current US Food and Drug Administration (FDA)-approved commercial NT-proBNP assays are based on Roche-manufactured monoclonal antibodies (mAbs) that are specific to the epitopes 27–31 and 42–46 amino acid residues (aar) in the central region of NT-proBNP. The 42–46 aar region of NT-proBNP includes Ser44, which can be modified by glycosidic residues. Several studies have shown that NT-proBNP assays underestimate the concentration of plasma NT-proBNP measuring only a subfraction of the circulating NT-proBNP because of the negative effect of glycosylation on NT-proBNP recognition by the antibodies [4], [5], [6].
The NT-proBNP concentrations measured by the Elecsys proBNP II assay (Roche Diagnostics) were shown to be higher and the diagnostic and prognostic accuracy were slightly improved by pretreating plasma samples with deglycosylation enzymes [6]. However, the use of glycosidases is time consuming and expensive, and may not be feasible in the clinical setting. An example of a cheaper and less laborious method for measuring total NT-proBNP is the SuperFlex NT-proBNP assay developed by PerkinElmer. This assay utilizes two mAbs specific to non-glycosylated regions of NT-proBNP located in the N- and C-terminal parts of the molecule. The assay is able to measure the total NT-proBNP without requirement of any special pretreatment of the samples [4].
The present study aimed to compare the diagnostic accuracy of the total NT-proBNP measured by the SuperFlex NT-proBNP assay with that of the subfraction of NT-proBNP recognized by the Elecsys Roche proBNP II assay in distinguishing patients with HF from individuals without HF.
Materials and methods
Study population
The present study carried out at Wuhan Asia Heart hospital (Wuhan, Hubei province, China) was approved by the Ethics Committee in accordance with the Helsinki Declaration. We recruited 383 consecutive patients with suspected HF for inpatient treatment from June 2019 to May 2021. The mean age was 63.2 years (±10.8 years); 165 patients (43%) were females. The detailed clinical characteristics of the study population are summarized in Table 1.
Descriptive statistics for the study population.
| HF | Non-HF | p-Value | |
|---|---|---|---|
| n (%) | 183 (47.8) | 200 (52.2) | |
| Age, years, mean ± SD | 65 (±10.8) | 61.5 (±10.9) | 0.002 |
| Female, n (%) | 75 (41) | 85 (43) | 0.764 |
| Body mass index, kg/m2, mean ± SD | 24.5 (4.1) | 25.1 (3.1) | 0.099 |
| LVEF | |||
| Normal (LVEF >50%), n (%) | 33 (18) | 160 (80) | <0.001 |
| Mild impairment (LVEF 40–50%), n (%) | 72 (39.3) | 16 (8) | <0.001 |
| Moderate impairment (LVEF 30–40%), n (%) | 38 (20.8) | 4 (2) | <0.001 |
| Severe impairment (LVEF <50%), n (%) | 37 (20.2) | 0 (0) | <0.001 |
| N/A, n (%) | 3 (1.6) | 20 (10) | <0.001 |
| Reduced ventricular systolic function, n (%) | 116 (63.4) | 11 (5.5) | <0.001 |
| Reduced ventricular diastolic function, n (%) | 115 (62.8) | 149 (74.5) | 0.014 |
| Cardiac enlargement, n (%) | 123 (67.2) | 76 (38) | <0.001 |
| Wall motion abnormalities, n (%) | 90 (49.2) | 9 (4.5) | <0.001 |
| Valvular heart disease, n (%) | 132 (72.1) | 81 (40.5) | <0.001 |
| Diabetes, n (%) | 45 (24.6) | 74 (37) | 0.009 |
| Hypertension, n (%) | 109 (59.6) | 121 (53.5) | 0.852 |
| Coronary artery disease, n (%) | 38 (20.8) | 35 (17.5) | 0.416 |
| Acute coronary syndrome, n (%) | 64 (35) | 56 (28) | 0.142 |
| Kidney disease, n (%) | 61 (33.3) | 20 (10) | <0.001 |
| Myocardial infarction, n (%) | 21 (11.5) | 10 (5) | 0.020 |
| Atrial fibrillation, n (%) | 69 (37.7) | 19 (9.5) | <0.001 |
| Cardiomyopathy, n (%) | 73 (39.9) | 6 (3) | <0.001 |
| Arrhythmia, n (%) | 39 (21.3) | 37 (18.5) | 0.491 |
| Hyperlipidemia, n (%) | 13 (3.4) | 56 (14.6) | <0.001 |
| Pulmonary disease, n (%) | 52 (28.4) | 7 (3.5) | <0.001 |
| Thyroid disease, n (%) | 12 (6.6) | 7 (3.5) | 0.026 |
| Creatinine clearance, μmol/L, median (Q1–Q3) | 91 (76–109) | 75 (63–87) | <0.001 |
| eGFR, mL/min/1.73 m2, median (Q1–Q3) | 69 (50–87) | 89 (75–98) | <0.001 |
-
LVEF, left ventricular ejection fraction.
Study participants were classified into one of the following two categories: (a) non-HF patients (n=200) and (b) HF patients (n=183). This classification was done by two experienced cardiologists blinded to the NT-proBNP results and was based on a detailed history with evaluation of the patients’ medical reports, appropriate risk factor assessment, physical examination, 12-lead electrocardiography, chest radiography, two-dimensional echocardiography and an ongoing assessment of the patients’ clinical status. The two reviewers agreed on diagnosis for all patients. The inclusion criteria of the HF patients were based on the algorithm proposed by the Chinese guidelines for the diagnosis and treatment of heart failure requiring typical signs and symptoms of HF and objective evidence of structural or functional myocardial abnormality. In patients with left ventricular ejection fraction (LVEF) ≥50%, diagnosis of HF was based on symptoms and clinical signs of HF combined with echocardiographic indices associated with diastolic dysfunction (e.g., left atrial enlargement, LV hypertrophy, indices of LV relaxation abnormality) or other evidence of impaired myocardial function.
Blood for measurement of NT-proBNP concentrations was collected in the first 24 h after admission in vacutainer tube containing lithium-heparin as anticoagulant. Plasma was prepared within 30 min after blood collection by centrifugation at 4,000 Relative Centrifugal Force for 8 min at room temperature. All samples were stored at −80 °C until analysis. Creatinine was measured on hospital admission by an enzymatic method using Randox reagents (Randox, Mauguio, France), and estimated glomerular filtration rate was calculated using the Chronic Kidney Disease Epidemiology Collaboration creatinine equation.
NT-proBNP measurements
Samples of venous heparinized plasma were diluted 1:1 with 50 mM phosphate buffer (pH 6.0), mixed, and split into two Eppendorf tubes (0.5 mL per tube). A mixture containing 0.17 µL of O-glycosidase (New England Biolabs),1.5 µL of sialidase A (Prozyme) and 5.33 µL of 50 mmol/L phosphate buffer (pH 6.0) was added to obtain a final concentration of 150 mU/mL of O-glycosidase and 15 mU/mL sialidase A. 7 µL of 5 mmol/L phosphate buffer (pH 7) was added to the samples without glycosidases. The samples with glycosidases (treated samples) and without glycosidases (untreated samples) were incubated for 18 h at 37 °C, which has been reported to be sufficient to remove all sugar moieties from NT-proBNP without affecting the stability of the analyte [5, 6].
NT-proBNP was measured by the commercial sandwich electrochemiluminescence Elecsys proBNP II immunoassay using a Cobas e 602 analyzer (Roche Diagnostics) and by the SuperFlex NT-proBNP assay (PerkinElmer) according to the manufacturer’s instructions.
The Elecsys proBNP II NT-proBNP assay involved the simultaneous reaction of NT-proBNP in the patient sample with a biotinylated NT-proBNP mAb and a ruthenium-labeled NT-proBNP mAb. The fluorescence of the antigen-antibody complex was quantified. The calibration curve was generated using the proBNP II Calset (Roche Diagnostics). The total assay time was 18 min. The measuring range was up to 35,000 ng/L. The limit of detection (LoD) was 5 ng/L. The limit of quantitation at 20% coefficient of variation (CV) was 50 ng/L. Total CV was 4.6% at 44 ng/L, 2.6% at 126 ng/L, 1.8% at 2,410 ng/L and 3.8% at 33,606 ng/L.
The SuperFlex NT-proBNP assay by PerkinElmer is based on coat mAbs 15C4 specific to the region 63–71 of NT-proBNP and detection mAbs 18H5 specific to the region 15–20 of NT-proBNP labeled with acridinium ester (both mAbs from HyTest). The assay is performed using superparamagnetic microparticles together with direct chemiluminescence technology to detect NT-proBNP in serum, plasma, and venous whole blood specimens. First the antigen from the specimen binds to the anti-NT-proBNP mAb 15C4 coupled to magnetic particles. After washing, acridinium ester-labeled anti-NT-proBNP mAb 18H5 is added to form an immunocomplex. Unbound substances are removed by washing and the luminescence value of the chemiluminescence reaction is measured under the action of pre-trigger and trigger solution. The luminous intensity is positively correlated with the concentration of the NT-proBNP in the specimen.
The assay shows test results on the reader display within 10 min of sample application. The measuring range is up to 30,000 ng/L. The LoD was determined to be 15 ng/L. The limits of quantitation in whole blood at 20 and 10% CV were <9 ng/L and <30 ng/L, respectively. Total CV was 2.4% at 35 ng/L and 2.0% at 140 ng/L. In a general population of 255 healthy subjects (mean age [±SD] 44.7 [±12.5], 34% female) median (Q1–Q3) NT-proBNP levels were 28 (17–59) ng/L. Recombinant non-glycosylated NT-proBNP produced in E. coli (HyTest) was used as a calibrator.
Statistical analysis
Categorical variables are expressed as the absolute number and percentages. Quantitative variables are described as mean ± standard deviation (SD) in the case of normal distribution or median (interquartile range) otherwise. To compare between HF and non-HF groups characteristics and clinical parameters, the unpaired t-test was used for continuous variables with a normal distribution, and the Mann–Whitney U test for continuous variables without a normal distribution, the two-proportion Z test was used to compare proportions. Correlations were assessed by the Pearson method. NT-proBNP concentrations below the limit of detection were substituted with the lowest limit of detection value. Diagnostic accuracy was assessed by ROC analysis with area under the curve (AUC) presented with 95% CIs. The accuracy was calculated as described by Bamber [7]. The covariance matrix of the AUCs was calculated as described by Delong and Delong [8]. To determine if ROC curves were statistically different from each other, the variance of the difference between two AUCs was determined and the p-value calculated assuming normality. p-Values<0.05 were considered statistically significant. Only concentrations above the limit of detection were considered for calculation of the percentage of non-glycosylated NT-proBNP in HF and non-HF patients. Statistical analyses were performed using XLSTAT statistical package by Microsoft®.
Results
SuperFlex NT-proBNP assay and NT-proBNP glycosylation
As the SuperFlex NT-proBNP assay is based on antibodies targeting non-glycosylated regions of NT-proBNP, we expected to see no or little effect of deglycosylation with enzymes on NT-proBNP concentration. NT-proBNP was measured in 90 heparin plasma samples (both from HF and non-HF patients) either treated or untreated with a mixture of glycosidases. We detected no effect of the glycosidase treatment on NT-proBNP values measured by the SuperFlex NT-proBNP assay, as NT-proBNP concentrations in treated and non-treated samples were almost identical (slope of 1.002) and highly correlated (r=0.99, p<0.001) (Figure 1).
NT-proBNP concentrations measured by the SuperFlex NT-proBNP assay. Scatter plot of NT-proBNP concentrations measured in the untreated and glycosidase-treated samples.
Comparison of NT-proBNP and total NT-proBNP concentrations
NT-proBNP was measured in both untreated plasma samples and those treated with a mixture of glycosidases by the proBNP II assay and in untreated samples by the SuperFlex NT-proBNP assay.
Concentrations were markedly higher in the glycosidase treated samples compared to untreated samples (p<0.001) in case of the proBNP II assay and were comparable to those in the untreated samples measured by the SuperFlex NT-proBNP in both HF and non-HF groups (p=0.546 and p=0.789, respectively) (Table 2).
NT-proBNP concentrations.
| Assay | Median NT-proBNP (Q1–Q3), ng/L | p-Value | |
|---|---|---|---|
| Non-HF | HF | ||
| ProBNP II w/o glycosidases | 85 (24–234) | 2,724 (1,037–6,242) | <0.001 |
| ProBNP II with glycosidases | 301 (146–749) | 5,850 (2,348–13,999) | <0.001 |
| SuperFlex NT-proBNP w/o glycosidases | 315 (158–735) | 4,954 (1,966–12,846) | <0.001 |
The measurements in the untreated samples by the proBNP II assay correlated with the total NT-proBNP concentrations measured in the glycosidase-treated samples (r=0.94, p<0.001) and with the measurements in the untreated samples by the SuperFlex NT-proBNP assay (r=0.92, p<0.001) (Figure 2A and B).
NT-proBNP concentrations measured by the proBNP II assay and SuperFlex NT-proBNP assay. (A) Scatter plot of NT-proBNP concentrations measured by the proBNP II assay in untreated and glycosidase-treated plasma samples. (B) Scatter plot of NT-proBNP concentrations in untreated plasma samples measured by the SuperFlex NT-proBNP assay and by the proBNP II assay. (C) Scatter plot of NT-proBNP concentrations in the untreated plasma samples measured by the SuperFlex NT-proBNP assay and in the glycosidase-treated samples measured by the proBNP II assay.
The total NT-proBNP concentrations measured in the untreated samples by the SuperFlex NT-proBNP assay showed a very close correlation with the total NT-proBNP concentrations measured by the proBNP II assay in the glycosidase-treated samples (r=0.99, p<0.001) (Figure 2C).
Diagnostic performance of NT-proBNP and total NT-proBNP
ROC curves for the diagnosis of HF in the whole study population are shown in Figure 3. The AUC to discriminate HF from non-HF patients was 0.943 (95% CI 0.922–0.964) for NT-proBNP measured in untreated samples by the proBNP II assay and 0.935 (0.913–0.958) for total NT-proBNP measured in glycosidase-treated samples by the proBNP II assay. There was no significant difference between the ROC AUCs of NT-proBNP measured by the proBNP II assay in untreated samples vs. glycosidase-treated samples (p=0.0568). The optimal cut-off values were 630 ng/L and 1,296 ng/L for NT-proBNP measured in untreated and glycosidase-treated samples (with accuracy of 88 and 87%, respectively).
ROC curves for diagnosis of HF in the whole study population.
The ROC AUC of total NT-proBNP measured by the SuperFlex NT-proBNP assay in untreated samples was slightly lower compared with that by the proBNP II assay 0.930 (95% CI 0.907–0.954) vs. 0.943 (95% CI 0.922–0.964) (p=0.0128). The optimal cut-off value was 1,240 ng/L (with accuracy of 87%).
Glycosylation of NT-proBNP in non-HF vs. HF patients
We estimated the percentage of non-glycosylated NT-proBNP at Ser44 in HF and non-HF patients as: (non-glycosylated NT-proBNP/[total NT-proBNP]) × 100%. The measurements of the untreated samples by the proBNP II assay were taken as the concentration of the non-glycosylated NT-proBNP at Ser44. We compared the results of our calculations based on total NT-proBNP measurements obtained either by the proBNP II assay in case of the glycosidase-treated samples or by the SuperFlex NT-proBNP assay in case of the untreated samples. As shown in Figure 4, the % of non-glycosylated NT-proBNP was significantly lower in non-HF patients compared with that in HF patients (p<0.001), suggesting that the proBNP II assay underestimates NT-proBNP to a greater extent in non-HF patients than in HF patients due to the higher level of NT-proBNP glycosylation in the non-HF group. The ratio of non-glycosylated NT-proBNP to total NT-proBNP based on the results obtained by the proBNP II assay in glycosidase-treated samples and by the SuperFlex NT-proBNP assay in untreated samples was 1.54 and 1.61, respectively (p<0.001).
Box plots showing mean percentage of non-glycosylated NT-proBNP in non-HF and HF patients. (A) The percentage of non-glycosylated NT-proBNP was calculated based on the results of NT-proBNP measurements by the proBNP II assay in untreated and glycosidase-treated samples. Median % (Q1–Q3): 28.7 (21.0–37.9) in HF group vs. 43.7 (36.1–53.5) in non-HF group, p<0.001. (B) The percentage of non-glycosylated NT-proBNP was calculated based on the results of NT-proBNP measurements by the SuperFlex NT-proBNP assay and by the proBNP II assay in untreated samples. Median % (Q1–Q3): 30.3 (21.3–42.1) in HF group vs. 49.0 (38.1–60.2) in non-HF group, p<0.001.
Discussion
The measurement of BNP and NT-proBNP by immunoassays has become a widely used tool for diagnosing HF. BNP assays are heterogeneous and measure both the multiple breakdown products of BNP and the precursor of BNP, proBNP [9], [10], [11]. In contrast to BNP assays, all current US FDA-approved NT-proBNP assays are based on the same antibodies and calibrator materials distributed by Roche [9, 12].
Circulating NT-proBNP and proBNP are O-glycosylated [13], [14], [15]. NT-proBNP assays mainly reflect the NT-proBNP concentrations, because circulating NT-proBNP concentrations are several-fold higher compared with proBNP concentrations [16, 17]. Commercial NT-proBNP assays underestimate the concentration of circulating NT-proBNP (in some patients up to 8-14-fold) because of the negative effect of glycosylation on NT-proBNP recognition by the antibodies specific to the middle fragment of the molecule [4], [5], [6].
Studies that linked glycosylation of NT-proBNP to diagnosis or clinical outcomes are still limited. Røsjø et al. showed that the deglycosylation of NT-proBNP by pretreatment of plasma samples with a mix of O-glycosidases slightly improved the diagnostic and prognostic accuracy of the Elecsys proBNP II assay. They measured NT-proBNP in plasma samples collected in standard EDTA tubes and total NT-proBNP concentrations in EDTA tubes with added glycosidases in unselected patients with dyspnea. Although both NT-proBNP and total NT-proBNP were predictive of death in patients with HF at a median of 814 days follow-up, the total NT-proBNP concentration was stronger in predicting death than the NT-proBNP subfraction [6].
Understandably, the treatment of plasma samples with glycosidases before NT-proBNP measurements is unlikely to be feasible in a clinical setting. For clinical use of the total NT-proBNP as a biomarker, a simple, high-throughput, and robust method is required. Considering the known sensitivity of antibodies to the presence of O-glycans in the recognized epitopes, total NT-proBNP can be measured by means of an immunoassay based on antibodies targeting NT-proBNP regions that are not modified by O-glycans. According to Seferian et al., N-terminal (13–24 aar) and the C-terminal portion of the molecule (61–76 aar) are mostly free of O-glycans and are suitable for this purpose [4].
The SuperFlex NT-proBNP assay developed by PerkinElmer is an example of an NT-proBNP assay that is able to detect total NT-proBNP without any requirement of special pretreatment of samples. This assay utilizes NT-proBNP antibodies specific to the regions 15–20 and 63–71 aar of NT-proBNP molecule. The assay is currently in clinical use in China and was approved by the national medical products administration.
In this study, we evaluated the diagnostic utility of the SuperFlex NT-proBNP assay in comparison with the proBNP II assay. Measurements of total NT-proBNP by the proBNP II assay in samples treated with a mixture of O-glycosidases was used as a method that has already been validated for detection of total NT-proBNP in plasma samples [5, 6].
The results of NT-proBNP measurements by the SuperFlex NT-proBNP assay showed a very close correlation with the results of NT-proBNP measurements by the proBNP II assay in samples treated with glycosidases, suggesting the reliability of this approach for measuring the total NT-proBNP level. In accordance with previously published results, we observed that despite nearly two-fold higher values there were no significant differences in the diagnostic performance of total NT-proBNP compared to that of a subfraction of NT-proBNP measured by the proBNP II assay for distinguishing HF from non-HF patients [6, 18].
The lack of substantial difference in diagnostic performance between the two assays support continued use of the established NT-proBNP commercial assay in the diagnosis of HF. However, a markedly lower percentage (1.5–1.6-fold) of non-glycosylated NT-proBNP to total NT-proBNP in non-HF patients compared to that in patients with HF suggests a potential difference in clinical value between the two assays in certain groups of patients. The impact of underestimation of NT-proBNP concentrations due to glycosylation may be more pronounced in patients without current or prior symptoms of HF or with mild HF than in those with pronounced HF conditions. Intriguingly, different degree of glycosylation at Thr71 of proBNP in different HF etiologies has been previously shown, i.e., patients with acute decompensated HF (ADHF), non-ADHF and chronic HF patients [19]. Glycosylation at this site exhibits an inhibitory effect on the convertases that cleave proBNP and may regulate the efficiency of active BNP production [15, 20].
Taking into account the known high variability in levels and site occupancy of O-glycosylated proteins, one may expect that immunoassays measuring total NT-proBNP levels might be advantageous for HF diagnostics, prognostication, and/or therapy monitoring in certain groups of patients and disease states due to their ability to detect endogenous NT-proBNP independently of its glycosylation status. As previously shown, the use of NT-proBNP thresholds based on European reference ranges to assess cardiovascular risk in community care may lead to under-diagnosis of evolving HF in Pacific Islanders. The NT-proBNP levels in Pacific adults were two-fold lower compared to that in Europeans when measured by the Elecsys proBNP II assay [21]. This underestimation may be caused by glycosylation of antibody-binding sites as previously discussed.
The Elecsys proBNP II assay was developed before any knowledge of NT-proBNP/proBNP glycosylation. Clinically used cut-off points are based on this assay and have been optimized for acute, severe HF syndromes; however, lower concentrations of BNP and NT-proBNP have been shown to have associations with structural heart disease and risk of incident HF in individuals both unaffected by heart disease and those with higher risk, such as those with pre-HF-type phenotypes like diabetes mellitus and cardiovascular disease [22, 23]. As an assay measuring total NT-proBNP results in higher concentrations being measured, this could be considered advantageous in the non-HF setting such as population screening. Obviously, an assay for total NT-proBNP will require new cut-offs to be clinically validated.
Increasing evidence shows that NT-proBNP measurements can be used for risk classification of patients with acute coronary syndrome, in which a low level of NT-proBNP is correlated to a lower risk [24, 25]. The results of the recent study by Lyngbakken and colleagues reemphasize the potential value of BNP and NT-proBNP measurement in individuals completely unaffected by obvious HF with the potential opportunity to prevent HF [26]. In this view, total NT-proBNP concentration assessment might reflect the real condition of cardiac stress and be valuable to evaluate the long-term prognosis or risk for patients.
The present study has several limitations. It was a single-center study performed with a limited number of patients. Furthermore, there was some variation in blood collection timing, and the etiology of HF was not accounted for in our analyses. Further studies with external validation of our results in other cohorts and more exploration of the clinical value of total NT-proBNP in different subgroups of patients are warranted.
Funding source: HyTest group
Acknowledgments
We are grateful to Dr. Natalia N. Tamm and Dr. Alexey A. Konev for the constructive criticism and helpful comments in the preparation of this manuscript.
-
Research funding: This study was supported by HyTest group.
-
Author contributions: All authors have accepted responsibility for the entire content of this manuscript and approved its submission.
-
Competing interests: Employment or Leadership: A.G. Semenov, HyTest group; E.E. Feygina, HyTest group; C. Yang HyTest group; N. Wang HyTest group; C. Chen, HyTest group; X. Hu, PerkinElmer; X. Ni, PerkinElmer.
-
Informed consent: Informed consent was obtained from all individuals included in this study.
-
Ethical approval: The present study carried out at Wuhan Asia Heart hospital (Wuhan, Hubei province, China) was approved by the Ethics Committee in accordance with the Helsinki Declaration.
References
1. Ponikowski, P, Voors, AA, Anker, SD, Bueno, H, Cleland, JG, Coats, AJ, et al.. 2016 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure: the Task Force for the diagnosis and treatment of acute and chronic heart failure of the European Society of Cardiology (ESC). Developed with the special contribution of the Heart Failure Association (HFA) of the ESC. Eur J Heart Fail 2016;18:891–975. https://doi.org/10.1002/ejhf.592.Search in Google Scholar PubMed
2. Yancy, CW, Jessup, M, Bozkurt, B, Butler, J, Casey, DEJr., Colvin, MM, et al.. 2017 ACC/AHA/HFSA focused update of the 2013 ACCF/AHA guideline for the management of heart failure: a report of the American college of cardiology/American heart association task Force on clinical practice guidelines and the heart failure society of America. Circulation 2017;136:e137–61. https://doi.org/10.1016/j.jacc.2017.04.025.Search in Google Scholar PubMed
3. Kavsak, PA, Lam, CSP, Saenger, AK, Jaffe, AS, Collinson, P, Pulkki, K, et al.. Educational recommendations on selected analytical and clinical aspects of natriuretic peptides with a focus on heart failure: a report from the IFCC committee on clinical applications of cardiac bio-markers. Clin Chem 2019;65:1221–7. https://doi.org/10.1373/clinchem.2019.306621.Search in Google Scholar PubMed
4. Seferian, KR, Tamm, NN, Semenov, AG, Tolstaya, AA, Koshkina, EV, Krasnoselsky, MI, et al.. Immunodetection of glycosylated NT-proBNP circulating in human blood. Clin Chem 2008;54:866–73. https://doi.org/10.1373/clinchem.2007.100040.Search in Google Scholar PubMed
5. Nishikimi, T, Ikeda, M, Takeda, Y, Ishimitsu, T, Shibasaki, I, Fukuda, H, et al.. The effect of glycosylation on plasma N-terminal proBNP-76 levels in patients with heart or renal failure. Heart 2012;98:152–61. https://doi.org/10.1136/heartjnl-2011-300102.Search in Google Scholar PubMed
6. Rosjo, H, Dahl, MB, Jorgensen, M, Roysland, R, Brynildsen, J, Cataliotti, A, et al.. Influence of glycosylation on diagnostic and prognostic accuracy of N-terminal pro-B-type natriuretic peptide in acute dyspnea: data from the Akershus Cardiac Examination 2 Study. Clin Chem 2015;61:1087–97.10.1373/clinchem.2015.239673Search in Google Scholar PubMed
7. Bamber, D. The area above the ordinal dominance graph and the area below the receiver operating characteristic graph. J Math Psychol 1975;12:387–415. https://doi.org/10.1016/0022-2496(75)90001-2.Search in Google Scholar
8. DeLong, ER, DeLong, DM, Clarke-Pearson, DL. Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics 1988;44:837–45. https://doi.org/10.2307/2531595.Search in Google Scholar
9. Saenger, AK, Rodriguez-Fraga, O, Ler, R, Ordonez-Llanos, J, Jaffe, AS, Goetze, JP, et al.. Specificity of B-type natriuretic peptide assays: cross-reactivity with different BNP, NT-proBNP, and proBNP peptides. Clin Chem 2017;63:351–8. https://doi.org/10.1373/clinchem.2016.263749.Search in Google Scholar PubMed
10. Semenov, AG, Tamm, NN, Apple, FS, Schulz, KM, Love, SA, Ler, R, et al.. Searching for a BNP standard: glycosylated proBNP as a common calibrator enables improved comparability of commercial BNP immunoassays. Clin Biochem 2017;50:181–5. https://doi.org/10.1016/j.clinbiochem.2016.11.003.Search in Google Scholar PubMed
11. Feygina, EE, Artemieva, MM, Postnikov, AB, Tamm, NN, Bloshchitsyna, MN, Medvedeva, NA, et al.. Detection of neprilysin-derived BNP fragments in the circulation: possible insights for targeted neprilysin inhibition therapy for heart failure. Clin Chem 2019;65:1239–47. https://doi.org/10.1373/clinchem.2019.303438.Search in Google Scholar PubMed
12. Semenov, AG, Katrukha, AG. Analytical issues with natriuretic peptides – has this been overly simplified? Ejifcc 2016;27:189–207.Search in Google Scholar
13. Schellenberger, U, O’Rear, J, Guzzetta, A, Jue, RA, Protter, AA, Pollitt, NS. The precursor to B-type natriuretic peptide is an O-linked glycoprotein. Arch Biochem Biophys 2006;451:160–6. https://doi.org/10.1016/j.abb.2006.03.028.Search in Google Scholar PubMed
14. Liang, F, O’Rear, J, Schellenberger, U, Tai, L, Lasecki, M, Schreiner, GF, et al.. Evidence for functional heterogeneity of circulating B-type natriuretic peptide. J Am Coll Cardiol 2007;49:1071–8. https://doi.org/10.1016/j.jacc.2006.10.063.Search in Google Scholar PubMed
15. Semenov, AG, Postnikov, AB, Tamm, NN, Seferian, KR, Karpova, NS, Bloshchitsyna, MN, et al.. Processing of pro-brain natriuretic Peptide is suppressed by O-glycosylation in the region close to the cleavage site. Clin Chem 2009;55:489–98. https://doi.org/10.1373/clinchem.2008.113373.Search in Google Scholar PubMed
16. Tamm, NN, Seferian, KR, Semenov, AG, Mukharyamova, KS, Koshkina, EV, Krasnoselsky, MI, et al.. Novel immunoassay for quantification of brain natriuretic peptide and its precursor in human blood. Clin Chem 2008;54:1511–8. https://doi.org/10.1373/clinchem.2007.100545.Search in Google Scholar PubMed
17. Seferian, KR, Tamm, NN, Semenov, AG, Mukharyamova, KS, Tolstaya, AA, Koshkina, EV, et al.. The brain natriuretic peptide (BNP) precursor is the major immunoreactive form of BNP in patients with heart failure. Clin Chem 2007;53:866–73. https://doi.org/10.1373/clinchem.2006.076141.Search in Google Scholar PubMed
18. Mueller, T, Gegenhuber, A, Poelz, W, Haltmayer, M. Comparison of the Biomedica NT-proBNP enzyme immunoassay and the Roche NT-proBNP chemiluminescence immunoassay: implications for the prediction of symptomatic and asymptomatic structural heart disease. Clin Chem 2003;49:976–9. https://doi.org/10.1373/49.6.976.Search in Google Scholar PubMed
19. Vodovar, N, Seronde, MF, Laribi, S, Gayat, E, Lassus, J, Boukef, R, et al.. Post-translational modifications enhance NT-proBNP and BNP production in acute decompensated heart failure. Eur Heart J 2014;35:3434–41. https://doi.org/10.1093/eurheartj/ehu314.Search in Google Scholar PubMed
20. Nakagawa, Y, Nishikimi, T, Kuwahara, K, Fujishima, A, Oka, S, Tsutamoto, T, et al.. MiR30-GALNT1/2 axis-mediated glycosylation contributes to the increased secretion of inactive human prohormone for brain natriuretic peptide (proBNP) from failing hearts. J Am Heart Assoc 2017;6:e003601. https://doi.org/10.1161/JAHA.116.003601.Search in Google Scholar PubMed PubMed Central
21. Faatoese, A, Lasse, M, Lewis, L, Pemberton, CJ, Richards, AM, Cameron, VA. Plasma NT-proBNP concentrations are two-fold lower in pacific versus European population samples: evidence for glycosylation impacting assay detection and/or ProBNP processing. Circulation 2020;142:A16095. https://doi.org/10.1161/circ.142.suppl_3.16095.Search in Google Scholar
22. de Boer, RA, Nayor, M, deFilippi, CR, Enserro, D, Bhambhani, V, Kizer, JR, et al.. Association of cardiovascular biomarkers with incident heart failure with preserved and reduced ejection fraction. JAMA cardiol 2018;3:215–24. https://doi.org/10.1001/jamacardio.2017.4987.Search in Google Scholar PubMed PubMed Central
23. Januzzi, JLJr., Xu, J, Li, J, Shaw, W, Oh, R, Pfeifer, M, et al.. Effects of canagliflozin on amino-terminal pro-B-type natriuretic peptide: implications for Cardiovascular risk reduction. J Am Coll Cardiol 2020;76:2076–85. https://doi.org/10.1016/j.jacc.2020.09.004.Search in Google Scholar PubMed
24. Mathbout, M, Asfour, A, Leung, S, Lolay, G, Idris, A, Abdel-Latif, A, et al.. NT-proBNP level predicts extent of myonecrosis and clinical adverse outcomes in patients with ST-elevation myocardial infarction: a pilot study. Med Res Arch 2020;8:10. https://doi.org/10.18103/mra.v8i2.2039.Search in Google Scholar PubMed PubMed Central
25. Schellings, DA, Adiyaman, A, Dambrink, JE, Gosselink, AM, Kedhi, E, Roolvink, V, et al.. Predictive value of NT-proBNP for 30-day mortality in patients with non-ST-elevation acute coronary syndromes: a comparison with the GRACE and TIMI risk scores. Vasc Health Risk Manag 2016;12:471–6. https://doi.org/10.2147/vhrm.s117204.Search in Google Scholar
26. Lyngbakken, MN, Kvisvik, B, Aagaard, EN, Berge, T, Pervez, MO, Brynildsen, J, et al.. B-type natriuretic peptide is associated with indices of left ventricular dysfunction in healthy subjects from the general population: the akershus cardiac examination 1950 study. Clin Chem 2021;67:204–15. https://doi.org/10.1093/clinchem/hvaa257.Search in Google Scholar PubMed
© 2022 the author(s), published by De Gruyter, Berlin/Boston
This work is licensed under the Creative Commons Attribution 4.0 International License.
