Since the Food and Drug Administration (FDA) cleared the first B-type natriuretic peptide (BNP) assay in 2000, natriuretic peptide (NP) testing—including BNP, N-terminal proBNP (NT-proBNP), and mid-region-pro-atrial NP—has, in the words of esteemed cardiologist and researcher James Januzzi, MD, “revolutionized” heart failure (HF) care. Assessing levels of these neurohormones, which are released by the heart in response to myocardial wall stretch from increased volume and pressure load, has become essential in diagnosing HF earlier, aiding in determining HF severity, and providing valuable prognostic information for HF patients.
Even as influential guidelines in the U.S. and at least six other countries or regions have endorsed the NPs for these purposes, research continues into how they might be used to guide HF care. Vibrant NP-related research also has peeled back many layers of complexity about NP biochemistry as well as nuances in measuring these analytes. In addition, studies of the new HF medication approved by FDA in 2015, valsartan/sacubitril (Entresto), a combination angiotensin II receptor blocker-neprilysin inhibitor, has raised still other questions about interpreting NP test results.
“Part of our due diligence as a lab community is to be aware of these sorts of things,” said Allan S. Jaffe, MD, professor of cardiology and of laboratory medicine and pathology and chair of the division of clinical core laboratory services in the department of laboratory medicine and pathology at Mayo Clinic in Rochester, Minnesota. “Unfortunately, there can be a sense that all assays are the same and that we can believe in the ‘magic’ numbers.”
An Intricate Tableau
BNP is a product of the precursor protein, pro-BNP, which the heart produces continually under normal demands. When pro-BNP is cleaved enzymatically, it releases the active hormone BNP and an inactive fragment, NT-proBNP. Many different immunodetection-based platforms measure NPs, but the tests have not been harmonized and, especially in the case of BNP, use different antibodies and target different epitopes, which affect test results.
Adding to this intricate tableau, NPs also have significant biological variability. “The intra-individual variability of NPs is very great,” said Ellis Jacobs, PhD, DABCC, FACB, director of scientific affairs at Alere, which produces both BNP and NT-proBNP assays, the latter of which is not available in the U.S. “As you get older, NP levels naturally increase. As you get heavier, your NPs decrease significantly; both BNP and NT-proBNP. These are things doctors need to consider when looking at their patients and their test results.” He added that declining renal function also affects NPs—NT-proBNP disproportionally more than BNP.
NP analytic challenges mostly, but not exclusively, involve BNP assays. As Jaffe and his co-author Vlad Vasile, MD, PhD, describe in a review article in the January 2017 issue of Clinical Chemistry, all NT-proBNP assays—distributed by Roche—use the same antibodies and calibrators, making for more consistent results across platforms, with <10% variation across methods (Clin Chem 2016; doi.10.1373/clinchem.2016.254724). NT-proBNP also tends to be more stable ex vivo than BNP, which requires EDTA tubes and does not store well.
Even with these plusses, NT-proBNP immunoassays are not the be-all and end-all in assessing HF. Most notably, they cross-react with nonglycosylated proBNP and do not detect glycosylated NT-proBNP and proBNP peptides. The latter occurs because of O-glycans in the epitopes that the assays’ monoclonal antibodies recognize in the central region of NT-proBNP. This means NT-proBNP assays typically measure only a portion of circulating NT-proBNP, and may, in the words of Vasile and Jaffe, “underestimate the concentration of circulating biomarker considerably”—perhaps by as much as 10 times.
BNP: Variability Is the Key
In contrast to NT-proBNP tests, there is a remarkable diversity and much more variability in BNP immunoassays. Available from multiple manufacturers, these assays use different capture and detection antibodies and target different parts of the BNP peptide. Consequently, results may vary by as much as 50% across platforms, according to Jaffe. They also cross-react with proBNP, a clinically relevant issue because proBNP is more prevalent in HF patients, he added.
Assay manufacturers and guidelines recommend a cutoff <100 ng/L to exclude HF, but Jaffe cautioned about the subtleties of interpreting results close to this threshold. “The first clinical trial that used BNP in the diagnosis of heart failure established this value to optimize sensitivity and specificity. Most companies then harmonized their assays at 100 ng/L, so it appeared that the assays gave similar performances. However, they weren’t harmonized beyond that number, something a lot of clinicians don’t understand—the results diverge above and below that number.” Indeed, Canadian researcher Andrew Don-Wauchope, MB.BCh, MD, has argued that “individual change in NP level should be more useful than absolute cut-points or target levels” (Clin Biochem 2015;48:236–46).
Reference materials to calibrate BNP assays also are lacking. Recently, researchers using internal standards provided by manufacturers and external calibrators found up to 3.6-fold differences between five BNP immunoassays, with a mean between-assay coefficient of variation (CV) of 24.5–47.2% (Clin Biochem 2016;
doi:10.1016/j.clinbiochem.2016.11.003). However, they were able to reduce this variability markedly, to a mean CV of 14.8%, when they used glycosylated proBNP expressed in HEK cells as the common calibrator for all assays.
With these knocks against BNP assays, some have been wondering whether they are on the verge of being mothballed in favor of NT-proBNP tests, especially in light of the PARADIGM-HF trial, which was suspended early in 2014 because of strong evidence that Entresto significantly reduced cardiovascular complications in comparison to standard therapy.
PARADIGM-HF found initial increases in BNP—that normalized after about 8 months—and substantial decreases in NT-proBNP. Questions have been raised about whether this was due to the drug’s effect not only on BNP release but also perhaps on glycosylation, NP biochemistry, or the analytics of the single BNP assay used in the trial.
“The NT-proBNP goes down but that is not a sign of improved heart function—just a sign of the drug having an effect,” explained Jacobs. “Part of the reason why NT-proBNP goes down is that there is speculation that neprilysin may be affecting the degree of glycosylation of NT-proBNP. If glycosylation increases, even though NT-proBNP is present, it will not be seen by the assay. This has caused a lot of confusion and makes difficult the interpretation of BNP and NT-proBNP in patients on this medication.”
Clearly, knowing a patient is on this drug would aid labs in making their NP-related interpretative comments, according to David Ledden, PhD, principal key expert at Siemens Healthineers, which offers both BNP and NT-proBNP assays.
Jacobs added that research is underway to better delineate the effect of neprilysin on NP biochemistry and to measure Entresto’s effect with different BNP assays. Uptake of this relatively expensive new drug remains unclear, but regardless it has “opened the door to a new class of drugs,” he said. “This is the first change in pharmacologic therapy for heart failure in years, so even if this specific drug does not have great market penetration, newer drugs in the same class with the same basic action probably will come out.” Jaffe also pointed out that basic science data suggest that human BNP is resistant to neprilysin.
Even with the subtleties of NP interpretation introduced by Entresto, in many ways the quality specifications for BNP and NT-proBNP assays first published in 2005 by the International Federation of Clinical Chemistry and Laboratory Medicine Committee on Standardization of Markers of Cardiac Damage still hold, according to Jaffe, one of the authors of this document (Clin Chem 2005;51:486–93). These include using age- and sex-specific reference and medical decision limits, being aware of the “inherently high” NP biological variation, considering the clinical setting in which NPs are used, and whether the test has been ordered for diagnostic, prognostic, or therapeutic guidance.
In addition to closely assessing product inserts of the assays they use, laboratorians might do well to dust off NP-related guidelines. In a recent review, Don-Wauchope, a medical biochemist for LifeLabs in Toronto, found certain guidelines provide more commentary on diagnostic cutoff values than others (Clin Biochem 2016;49:8–15). “As with any test, it is important for natriuretic peptides to be interpreted in the context of the clinical setting. The onus is on labs to provide the right information,” he said. “And it is up to physicians to make sure they apply that information to the patient they are seeing.”
Freelance writer Joely Johnson Mork contributed to this article.