| Speaker |
Speech |
| Chris Driscoll (00:05): |
Good afternoon everyone, and thank you for joining us for a multidisciplinary conversation about natriuretic peptides with views from both the laboratory and cardiology. I'm Chris Driscoll, publisher here at Medical Laboratory Observer. Today's program is sponsored by Beckman Coulter. Thank you to our sponsor and to you, our audience for giving us your time and attention today. |
| (00:25): |
Before we get started, we have a few housekeeping details. To submit a question, please use the Q&A box to the left of your video window at any time. You do not have to wait until the end of the program. For technical issues, please press F5 to refresh. If that does not work, you can submit a question with your issue through the Q&A panel. At this time, I'd like to introduce Dr. Robert Christenson, Professor of pathology and of medical and research technology at the University of Maryland School of Medicine, and Dr. James Januzzi, cardiologist and Director of the Dennis and Marilyn Barry Fellowship of Cardiology Research at Massachusetts General Hospital. Gentlemen, welcome to the program. |
| Robert Christenson (01:04): |
Thank you very much. Good day to everyone. |
| Dr. James Januzzi (01:07): |
Thank you. |
| Chris Driscoll (01:07): |
This first question, Dr. Januzzi, this is for you. How do you use natriuretic peptides assays in your practice today? |
| Dr. James Januzzi (01:15): |
Yeah, thanks very much for the question, and if you asked me this question maybe 10 or 15 years ago, it would be a much more straightforward answer. We used to use natriuretic peptide assays to evaluate patients with acute dyspnea in the emergency department pretty much as the only indication or the only role for the test. But that has changed dramatically over the last decade and a half. |
| (01:42): |
Really now we use it pretty much across all facets of evaluation of risk in heart failure. So while we still use it heavily in the emergency department setting to help us to identify or exclude acute heart failure as the cause of shortness of breath in the emergency department. We also use NT-proBNP or BNP to assess risk in that setting, so we can use it to estimate prognosis for hospitalized patients. In addition, we can use it to judge adequacy of heart failure therapy in the course of acute heart failure treatment where a decline in inpatient NT-proBNP or BNP may identify patients that are responding well to treatment. |
| (02:27): |
Furthermore, we can use the same functionalities in the office setting, so we can use it to evaluate shortness of breath in the office for the possibility of heart failure. In addition, we monitor prognosis in our patients with established heart failure in the course of outpatient treatment. Now, if you ask me this question five years from now, I think the answer would be very different. And I'm curious to see how Dr. Christenson feels about the evolving role, not only in the patients with symptoms, but possibly as part of a risk assessment for people at risk for heart failure. |
| Chris Driscoll (03:04): |
Dr. Chris, any comment to that? To the answer? |
| Robert Christenson (03:08): |
Well, I think yeah, definitely, and I think that this will bring up running it in a more general population for risk assessment in asymptomatic individuals in the general public. I think we'll bring a lot of other issues in, like reference intervals for example, making sure we have reference intervals that are personalized for the various populations. So I think it will be a lot of interesting items to discuss when we come up with those points. |
| Chris Driscoll (03:40): |
Great. Okay. Thank you. This next question is for both of you. What considerations are a factored when deciding between BNP and NT-proBNP? Are there conditions you may choose one over the other? What are the benefits to having access to both NT-proBNP and BNP? So this is for both of you. |
| Dr. James Januzzi (03:58): |
This is all for Rob to answer because it's a very heavily analytical answer, right? |
| Robert Christenson (04:03): |
Right. Let's take a half a step back. I mean, whenever you would have two assays for very similar analyte in the lab, why would you do it? You have to have a pretty compelling reason to do that. One compelling reason might be that if your assay that you have on your large analyzer is they don't have a point of care option and you want to run this at point of care, so that might be one alternative, one reason that you'd want to do that with. I really believe that on balance, otherwise the risks or liabilities outweigh going down this route. |
| (04:49): |
The major issue is confusion. I think that especially, I mean I, in an academic environment, had been my whole career, and whenever you have a whole new set of residents, whole new ... we're always changing personnel, to keep everyone educated as to which assay to use in which situation is very, very complicated and could actually lead to harm in some cases. It's very difficult to harmonize enough of our assays the way they are, and if we think about two different assays, harmonizing those values is going to be extremely complicated. So I don't believe that unless there's a point of care type of scenario going on, that in general, having two assays for analyzing to give you similar information is compelling enough to go through all that. |
| Dr. James Januzzi (05:49): |
Yeah, I mean, the only thing I'd add is, and this is something I've learned from Dr. Christenson and his colleagues in the laboratory world, I encourage clinicians to think about what instrumentation your hospital runs. I mean, to a large extent determines which assays might be selected. So it's sort of like saying, "Well, my laptop computer is an Apple computer, but I want to run Microsoft in it," I mean it's not going to work that way very well. So if your hospital is a place that runs instruments that would favor a BNP as the choice, that's probably the assay that's going to be selected because there's a much broader perspective here. |
| (06:33): |
There are hundreds of other analytes that these instruments have to run. And so I think that that first and foremost, at least for hospital-based systems, often determines the choice. Have there been changes in the number of institutions running one versus the other? Sure, but that's just the way that the system works. And so I wouldn't say that we can necessarily say that one is better than the other. And so it comes down often to a collaborative decision with your laboratory director. |
| Robert Christenson (07:05): |
And I think this brings up a terribly important point, and that is one of when you are choosing your big box analyzer, for example, for use in the laboratory, make sure those clinicians know what the alternatives are and what the tests that are in there. Like for example, BNP versus NT-proBNP, the type of troponin, even the types of hormones that are offered on many of these platforms because they could really impact the practice of the folks in your institution. So I think it just brings up the point that communication between the specialties is critical. |
| Chris Driscoll (07:45): |
Great. All right. Thank you both. Dr. Januzzi, how often are patients diagnosed with HF in your office? And do most of them come via referral such as being diagnosed in the ED? |
| Dr. James Januzzi (07:55): |
Yeah, great question. And it's changing over time as our sensitivity for recognizing the diagnosis has increased. Heart failure begins months to years before it's diagnosed in our patients. The universal definition of heart failure recognizes what we refer to as so-called stage B heart failure, which is otherwise known as pre heart failure, which by definition lacks symptoms. But by the time a person begins to develop symptoms, the diagnosis has often been present for a long time and may be very subtle. |
| (08:33): |
And so yes, we diagnose it in the office, but at least traditionally it was unfortunately diagnosed more in the emergency department setting because a person eventually culminated in severe symptoms and decompensation requiring hospitalization. I would say it is still 90/10 emergency department office-based initial diagnosis. Now, that's a little different than how frequently we see people with known heart failure decompensating, where more and more we're shifting to the office-based setting for management of decompensated heart failure. |
| (09:13): |
So many institutions, including ours at the Massachusetts General Hospital Heart Center will see patients with known heart failure who previously would have gone to the emergency department with their signs and symptoms where we can manage them with intravenous diuretics and other therapies rather than referring to the emergency department setting. So it's changing. Still, for initial diagnosis it's mainly a hospitalized diagnosis, but with respect to management of chronic syndromes, we're more and more shifting to the office. |
| Chris Driscoll (09:47): |
Thank you. In addition to known heart failure, any other clinical presentations where it'll guide your use of the natriuretic peptides? |
| Dr. James Januzzi (09:57): |
Well, yeah. Thank you for asking that because I think it's important for clinicians to recognize that we have unfortunately put natriuretic peptides into a box and we think of them as a heart failure biomarker. But the truth of the matter is that the heart releases natriuretic peptides under situations of stress. Therefore, there are circumstances where patients will have clinical diagnoses such as valvular heart disease or atrial fibrillation, is another great example where their natriuretic peptide values will be elevated and it can frustrate clinicians because they see that and they think, "Well, this patient doesn't have heart failure." |
| (10:32): |
Remember that once natriuretic peptides are elevated, it identifies so-called stage B heart failure. In other words, it's identifying stress. And the reason why we know this is because when you follow patients such as those that I've just described, without the overt diagnosis of heart failure, valvular heart disease, AFib with a high NT-proBNP or BNP, they eventually develop symptoms. And so it really speaks to the fact that we can make a molecular diagnosis of heart failure even before the clinical diagnosis is evident. |
| Robert Christenson (11:08): |
Great. I wonder if I could ask Dr. Januzzi a quick question here because it's one that ... Would you put patients with hypertension in that class? At least some of them? It seems that is such a prevalent diagnosis, especially in the community we serve here in Baltimore that I just wonder if natriuretic peptides that are called for in that population. |
| Dr. James Januzzi (11:34): |
Yeah. Thank you, Rob, for asking that question. And the answer is yes. This is an example of everything old is new again. One of the earliest NT-proBNP studies showing that one could predict risk in hypertension actually was from the LIFE study. Early on, early 2000s, the investigators showed that individuals with left ventricular hypertrophy and hypertension with an elevated natriuretic peptide had a risk quite comparable to patients who had already established symptomatic heart failure for future hospitalization. |
| (12:09): |
Now, we fast-forward many years into the future to about a year or two ago, there was a paper out of the sprint study looking at hypertension management showing that in people with elevated natriuretic peptides, more aggressive blood pressure lowering was associated with an even greater reduction event rates. Conversely and interestingly and very controversially, people with a normal natriuretic peptide but elevated blood pressure did not seem to accrue as much benefit from aggressive blood pressure lowering, suggesting that the blood tests may identify people where we should really be doubling down on blood pressure control, whereas people with normal natriuretic peptides, yes, we want to get their blood pressure down, but it implies that you have some time before the risk for heart failure and other cardiovascular complications might be evident. |
| Robert Christenson (13:02): |
And those regular, do you treat those folks with the ACEs and ARBs, the normal sort of standard pre heart failure treatments? |
| Dr. James Januzzi (13:15): |
Absolutely. Yeah. So the nice thing about natriuretic peptides is they don't discriminate about therapies. Anything that unloads the heart and reduces neurohormonal activities going to reduce them. So if I see someone with LVH, an elevated NT-proBNP or BNP, so now they have stage B heart failure, I would treat them with an ACE inhibitor as probably or an ARB preferentially, and then think about other therapies that I might use. |
| Robert Christenson (13:41): |
Great. Well, thank you very much for that enrichment. |
| Chris Driscoll (13:45): |
Thank you. Dr. Januzzi, another question for you. What are the implications of test results when your lab's natriuretic peptide tests differ from the natriuretic peptide test type run in the patient records? For example, when the patient was seen in the ED, what was run historically? For example, if the patient has recently moved to your clinic from another, what do you do in that situation? |
| Dr. James Januzzi (14:04): |
Yeah, I'm going to phone a friend and ask the past president of the AACC to weigh in on this one. But what I would say is that there's no easy answer other than education. We used to back many years ago, say, "Just divide by 10, and that gives you a ballpark," but that's just really wrong. There are some people whose BNPs and NT-proBNPs are only a factor of fourfold different. Some are tenfold, some are sixfold. So there's really no easy conversion between the two assays. Rob, I'd love to hear your thoughts. |
| Robert Christenson (14:37): |
Boy, it's just, I mean, BNP basically you have historically, it's had basically one cutoff right there at 100 and all the BNP assays seem to converge at 100. They all intersect at 100, but then they go out from there because there really isn't much in the way of standardization or harmonization between the various assays. And so you compare that with the NT-proBNP, which has age cutoffs that are defined, a cutoff perhaps for rule out, which is around 300. And the values are not that different, 300 from what you might see with another BNP type of assay. So I think that this is why I answered the question the way I did about having both assays in the same institution. It really is complicated. Unless you really have a great educational and sustained educational program, it's going to lead to confusion. |
| Dr. James Januzzi (15:49): |
I wouldn't do it. |
| Chris Driscoll (15:49): |
Continuing on your mention of the age related cutoffs, what are the benefits and how they differ from single cutoffs? So this question is for both of you could share your perspectives on the age related cutoff. |
| Dr. James Januzzi (16:01): |
Sure, yeah. Why don't I kick off and then we'll hear what Rob has to say? So I'll start off by saying that our group at the Mass General came up with the age stratified cutoffs that we now use globally for NT-proBNP, but I will also say that BNP is no different. The two peptides come from a common precursor. So it's not that age somehow magically affects a fragment of the precursor peptide without affecting the other. They're both very heavily age dependent. It's just that with NT-proBNP, in the earliest studies we sought to improve sensitivity for younger patients, it improved specificity for older patients. |
| (16:46): |
And so we examined numerous variables that help to determine loss of sensitivity in younger and loss of specificity in older. And what we found was in fact age, not surprisingly, which sums up a number of different physiologies including kidney function, the accumulation of structural heart disease that may be present as we get older. Really determined to some extent the operating characteristics that we see in our patients across the age spectrum. |
| (17:20): |
We looked at kidney function as a standalone. We looked at presence of known structural heart disease. At the end of the day, the simplest and most elegant way to do this was with age. And so we partitioned populations based on common age categories. We chose the average age in the original PRIDE study of right around 75 years between 65 and 75 is sort of the middle category, and then came up with the age cutoffs that were optimized for each group of patients. |
| (17:53): |
Now, once again, what I would say is that there's no difference between NT-proBNP and BNP in terms of the effects of age. But the way that the investigators went with BNP is rather than using age stratified cutoffs, they actually don't use a single cutoff. With BNP they talk about from 100 to 500 as sort of what they refer to as the gray zone, such that above 500, you have a high enough positive predictive value to say it's very likely to be decompensated heart failure. It just so happens that the single strongest predictor of values between 100 and 500 with BNP is age right? That was published by the Breathing Not Proper investigators. So it's just an example of two different strategies for managing the same effect of comorbidities on the biomarker. |
| Robert Christenson (18:46): |
So absolutely you're asking the right person that question because we have Dr. Januzzi and his group to credit for the age stratification there. I would just add one thing. When we use the term cutoff, probably best to talk about rule in cutoff and rule out cutoff because I think there would be a single cutoff. I think Dr. Januzzi's group opined that a single cutoff at about around 300 would be good for rule out heart failure. So I would just be careful when we say that, that we classify it as either rule in or rule out. |
| Dr. James Januzzi (19:25): |
Yeah. Absolutely, Rob. Thank you for that. And what Dr. Christenson is referring to is reflecting the fact that working together with folks like Dr. Christenson from the lab, in our earliest studies, we thought of NT-proBNP is a continuous variable and rather than dichotomizing it around a single cutoff. And so we identified a threshold value at a very low range that helped us to exclude heart failure. And then along the journey of age, we identified the threshold values that identified risk for acute heart failure. |
| (20:00): |
With BNP, I think we sacrifice ease of use when we use 100 because 100 is neither optimized for positive predictive value or negative predictive value. The equivalent threshold for an NT-proBNP of 300, which gives 99% negative predictive value is a BNP that's probably lower, somewhere around between 50, 70 in that range. So it's just, once again, it really explains why Dr. Christenson and I have both sort of said institutions should choose one peptide and educate their clinicians to know how to use them because all of these different cutoffs can potentially confuse folks. |
| Robert Christenson (20:45): |
I do feel like these specifications though for NT-pro do have value because that 300 is a number that's easy to remember. And the way that the cutoffs, the rule in cutoffs are like double, 450, 900, 1,800 I guess for the different age categories. I think those are good, easy things to remember there rather than having to think about the other issues with that 100, which came right off the ROC curve for minimizing the number of false positives and false negatives. So anyway. |
| Dr. James Januzzi (21:22): |
Yeah, a bit of trivia actually, the analogy to 100 for BNP. If you looked at the original, I remember the day sitting in my office back in about 2003, Rob, looking at the ROC curve for NT-pro from our very first study. It was 900. And it's sort of interesting that 900 sort of sits right in the middle of that rubric of 450, 900 and 1,800. And so if put up against the wall and said, "Give me one cutoff like we use BNP." I would say, "900 for NT-pro," but I would definitely use the age stratification approach because for older folks, 900's too low, and for younger folks it's too high. |
| Chris Driscoll (22:07): |
I'm going to turn into a little bit of a clinical slant here. Next question. This is for both of you though, but what are the most common complaints that you receive from clinicians regarding the natriuretic peptide assays and how do you address those? |
| Dr. James Januzzi (22:21): |
Dr. Christenson, do you want to start? |
| Robert Christenson (22:23): |
Yeah, so I think most of our ... really, the assays seem to work very well, and we in our institution use NT-proBNP. We don't get a lot of questions. I think a lot of that, we had a cardiologist named Christie Philippe who was a great colleague and a better friend, he's a really easy person to work with. And basically he was able to do the education that really set up our interpretive comments and set up cardiology and the emergency medicine folks to contact about the interpretation. |
| (23:06): |
The other is that I think the way that it's been thought about is, I guess I think about these as vectors, things like biomarkers. It can send you one way or the other with a diagnosis or with assessment. And I think we can use the natriuretic peptides even more effectively in the future for guiding therapy and all these other implications. But I think with all the other tools that cardiologists have to assist them in diagnosis and management that the NT-proBNP is but one of them. It's not quite like a cornerstone for heart failure or whatever because of the specificity, because it is a marker of hemodynamic stress. Anything that causes hemodynamic stress is going to cause an increase in NT-proBNP. So anyway, life is good with NT-proBNP here at Maryland. |
| Dr. James Januzzi (24:06): |
It's the same with us at Mass General. The days of being assaulted in the office or in the hallway about a result that a clinician didn't understand, those days are over. I mean, people have really learned how to use natriuretic peptides and understand very well that it's not just a biomarker of heart failure. I mean, I think we need to be really cautious about putting it in that box, in that box only. It's a biomarker of myocardial stress. |
| (24:37): |
And in fact, there are a lot of non heart failure indications where elevated natriuretic peptides may be useful for making decisions. Pulmonary thromboembolism is an example. Our clinicians use NT-proBNP to decide on thrombolytic therapy. Gram-negative sepsis, market elevation of NT-proBNP in individuals with shock from bacterial infection identifies individuals very likely to progress onto ARDS and require ICU level care. So when we think about the potential applications here, they really are endless. |
| (25:12): |
It's really our lack of imagination for how to utilize this test in a clinically structured fashion that has led to over-the-years confusion about what it's telling us. But since we've been measuring it now, gosh, I mean, Rob, it's what? 20 years at this point. We're really getting to the point where we understand better how to really utilize the test in a clinically logical manner. |
| Robert Christenson (25:37): |
Seems like just yesterday. It's amazing. But I also think that that's why we need to ... exactly what you're saying is why we have to sort of avoid calling it a false positive because it does confer some risk of pulmonary embolism and so on like the other issues you talked about, there is something going on. It may not be heart failure. In fact, probably about 40% of the time it's not according to some of the data, but it's something that one should have an eye on as either a harbinger, like with hypertension for example or another. So I try not to call them false positives, but anyhow, that's only my editorialization. |
| Chris Driscoll (26:29): |
Okay. Another question for both of you. When you're interpreting natriuretic peptide values, what are some key considerations that you feel are underappreciated? |
| Dr. James Januzzi (26:39): |
Underappreciated? Well, again, I think that our clinicians have gotten to understand some of the quirks and there are quirks. But why don't we start ... why don't I stick with some of the pre-analytical things? And Rob, I'd love to hear your thoughts as well. So pre analytically, in other words, for the clinicians, these are the factors before the blood is loaded on the machine, you should be thinking about in terms of how it might influence the result. And there are factors that relate to how the biomarker responds to being in a tube, et cetera, and so on. I'll let Rob speak to that. But there are clinical variables. |
| (27:29): |
We talked about age of course, and there are things besides wisdom and good looks that come with age, and that includes an accumulation of cardiac abnormalities that may or may not equal symptomatic heart failure, but may nudge the natriuretic peptide values higher. Kidney function is important. Probably one of the most misunderstood variables with respect to cardiac biomarkers. I think this one still, we need a lot more education on. Dr. Christenson and Dr. [inaudible 00:28:01], who he mentioned earlier, have done some landmark work trying to show the world that although yes, natriuretic peptides are cleared by the kidneys, they're small molecular weight proteins, and therefore there are accumulation should not be expected until the GFR is down under 15 or so. |
| (28:20): |
And so when you see people with rising natriuretic peptides in the setting of chronic kidney disease, it's more likely to be a true positive than a false positive on the basis of kidney dysfunction. Now, why? It's a long discussion, but kidney disease creates heart disease, and I think we can leave it at that and just basically make the point that one should not ignore a higher value in somebody with chronic kidney disease. |
| (28:46): |
I think it's also worth noting that there are factors that lower natriuretic peptides below a value you might expect. In other words, somebody comes in with class four heart failure and their NT-proBNP may be "not positive". Two thoughts on that. The first is there are clearly syndromes that are associated with lower than expected values. Heart failure with preserved ejection fraction, for example, owing to the fact that wall stress is lower in HFpEF. But obesity is another great example of lower than expected values. |
| (29:23): |
Dr. Christenson and team many years ago did a wonderful study showing that if you took individuals that higher body mass index and measured their transmural wall stress, in fact, those people with obesity actually had more wall stress than people who were not suffering from obesity, and yet their NT-proBNP was lower than expected. So why might that be? There's been a lot of work over the years. There's still a lot of discussion about it, but it turns out that adipose tissue, among other things, adipose tissue is active from a neurohormonal perspective, secreting a number of compounds including androgens, which may suppress the BNP gene. And we know this because there are actually now studies showing in many different ways blockade of such androgens from adipose tissue may be associated with the rise in NT-proBNP in patients with obesity. So that's just a few examples of things that clinicians should think about. But Rob, I'd love to hear your perspective as well. |
| Robert Christenson (30:25): |
Yeah, just a couple of things. I think on the pre-analytic piece of it, I think that NT-proBNP has been shown to be exquisitely stable through freezing and through storage before measurement and so on. I think the BNP has been shown to be much more labile. And so as far as a sendout lab where we've all done it, we all walk out of the office, we see these boxes for one of our brethren that are in the commercial lab business, and we say, "Oh my God, I wonder how hot it gets inside that thing and how that impacts the tubes of blood that are stored there?" |
| (31:11): |
So I think that one thing is stability. Using a type of tube, either heparinized or EDTA or serum that has been checked out by the FDA and has been validated, that's always a good idea unless you're willing to do those proper studies yourself. I think biologically or physiologically, a big issue that is an unmet need is sex with the natriuretic peptides, that women have higher values. The NHANES study that was just published not too long ago and looked at a large population of men and women, it seems by the time that women get up to the golden years, what happens is that 20% are over the upper reference limit. So this will become more and more important as these biomarkers move into use in the general population, having those reference intervals set at the right point. |
| (32:26): |
Just like a lot of things like creatinine, which renal function decreases with age, I think the natriuretic peptides with increases, especially in women. So that would be maybe something we should look at with the next generation of assays is maybe more sex specific or they take into our maybe multifactorial type of personalization of the values, especially if we want to use this in general medicine. |
| Dr. James Januzzi (33:00): |
Yeah, for wellness evaluation. Yeah. It's a really important point, Rob. The data that Dr. Christenson speaks of are spectacular out of NHANES that really show across the age range in the population the expected values. And it circles back to this issue of androgen based suppression of natriuretic peptides. One of the reasons why women have higher values is actually thought better as men have lower values because of the effects of androgen to reduce natriuretic peptides. |
| (33:32): |
And again, the reason why we know this is numerous examples now exist from various therapies to reduce androgen concentrations in men, for example, undergoing cancer chemotherapy for prostate cancer, where natriuretic peptide values rise small amounts, but nonetheless, significantly really just illustrating there are some important physiologic differences between men and women that need to be considered when we're looking down in these relatively normal ranges. |
| Robert Christenson (34:02): |
It's important to remember that this is really reflecting a hormone. NT-proBNP or BNP is really a heart hormone, and we know the values, the day-to-day values can vary quite a bit and actually substantially. So I guess it's not surprising that there would be a sex difference, but anyway. |
| Chris Driscoll (34:24): |
Thank you. Actually, in your answer, you actually kind of started to touch upon my next one, which is about the interpretation of results and how comorbidities are in play. Anything to add to that or do you feel like you've kind of covered where you wanted to go? |
| Dr. James Januzzi (34:40): |
I mean, I think Dr. Christenson and I have covered most of it. I will say that one of the most important things about the so-called "false negative". Remember, these cutoffs were developed to try to harness the information from a circulating blood hormone. So don't blame the blood test. More often than not, there's a reason for why the value may not be "positive", but maybe one of the most empowering things I tell my colleagues about interpretation of natriuretic peptide values is don't be seduced by the cutoff as being gospel. |
| (35:16): |
And so if somebody comes in with shortness of breath, they have obesity, they may have heart failure with preserved ejection fraction and their NT-proBNP is, I don't know, 600. That's not positive by the lab information system, but it's not normal. And so understanding that a normal NT-proBNP ... all right, the three of us, if we measured our NT-proBNP, 20, 30, something like that. So if you see a value that's in the three digits, certainly if you see a value that's above 300, it's more likely to be heart failure than not if the symptoms line up. It's all about the clinical picture that I think we need to consider here. |
| Robert Christenson (36:02): |
That's why also resisting that, oh, it's a false positive for heart failure. Well, it means something is going on when you get up into about three digits, as Dr. Januzzi says. |
| Chris Driscoll (36:14): |
Okay. Speaking of the peptide values, what if any medications or supplements may impact natriuretic peptide values? |
| Robert Christenson (36:26): |
Well, I think that's definitely in Dr. [inaudible 00:36:29]'s report. It's amazing to me how the ... in the 28 years, I guess, since NT-pro and BNP became widely available, the treatment and the management of heart failure has changed so much. Here, we've worked so hard to keep our biomarker exactly the same as it was when it was FDA cleared back then and gives the same values and all that. But meanwhile, on the clinical side, there's been so many advancements. So I'm looking forward to hearing about the impact of therapies. |
| Dr. James Januzzi (37:08): |
Sure. Yeah, I mean, it is really fascinating how despite all the advances in treatment when a person is sick, they're sick, and the cutoffs remain largely the same as they were in the early 2000s, despite all the treatment differences that have come about. The simple answer is that with a couple of exceptions, the therapies, whether we talk drug therapies or device therapies that are favorable for heart failure tend to lower natriuretic peptide values. |
| (37:46): |
Now, I think one of the mistakes that we made when that observation was first made was that we sort of thought, "Okay, great. We have a barometer for deciding on therapy intensity." And that's not really, I think, how to think about it anymore because one of the things that we forgot to ask is why do they go down? And what has been elucidated over the last few years is that secular trends in natriuretic peptides, whether you look at single values separated in time, or even better if you look at the longitudinal change over time with multiple measurements, are actually very good predictors of change in cardiac structure and function. |
| (38:32): |
So in people with a reduced ejection fraction therapies for heart failure, which tend to lower natriuretic peptides typically do so in exchange for improvements in ejection fraction. And so it's a useful way for clinicians as they measure serially in the office to utilize that that result more than just as my patient's doing well or they're not doing well. Prognostic biomarkers are useful, but they're really useful if you understand why they're prognostic. And so that's how the thinking has evolved about the role of how to interpret these tests over time. |
| (39:12): |
Now, not every treatment lowers natriuretic peptides equally. There are some that have a subtle effect. Beta blockers, for example, will cause an initial increase, but then a gradual drop over time as the heart reverse remodels and function improve. Diuretics when we decongest patients, certainly cause a reduction. The one pharmaceutical that has brought a lot of attention to the natriuretic peptides is sacubitril/valsartan, which is a combination drug with an angiotensin receptor blocker. |
| (39:50): |
And we know that ACE inhibitors and ARBs lower natriuretic peptides, but sacubitril/valsartan also includes a neprilysin inhibitor, and neprilysin is responsible for the breakdown of bioactive natriuretic peptides. So it should come as no surprise that since BNP is degraded by neprilysin, that when one goes on this drug, sacubitril/valsartan, there's a subtle impact on BNP values. In some studies, it would rise by about 20%. In our hands, actually, we've published a few studies now showing it doesn't rise, but it doesn't fall. |
| (40:28): |
Whereas NT-proBNP plummets by 30% to 40% when a person starts on this drug. And those reductions in NT-proBNP are very strongly related to the benefit of the agent in terms of reverse remodeling health status and outcomes. And so there is a bit of a dichotomy there. And with the inexorable rise of the use of this drug, it has caused a lot of people to essentially move away from frequent monitoring of BNP and people taking it because the results are a little bit more challenging to interpret. |
| Chris Driscoll (41:03): |
Thank you. And Dr. Januzzi, how do you know when or not to prescribe ARNIs? And does that impact the natriuretic peptide test you order? |
| Dr. James Januzzi (41:10): |
Yeah, that's the drug I'm talking about. It's angiotensin receptor. |
| Chris Driscoll (41:13): |
Yeah, that's it? |
| Dr. James Januzzi (41:14): |
Yeah. |
| Chris Driscoll (41:14): |
Okay. |
| Dr. James Januzzi (41:14): |
It's an angiotensin receptor, neprilysin inhibitor. I will say very quickly as an editorial comment that if you have BNP and that's all you have, it is still very prognostic when people are taking sacubitril/valsartan. So I don't want to be perceived as saying you can't measure it or you shouldn't measure it. You can. It's just that the secular trends over time are somewhat obscured by the pharmacologic rise in BNP related to the neprilysin inhibition combined with the prognostically favorable reduction that is also occurring. So this balance of increase in reduction often equals no change, whereas NT-proBNP goes down because it's not degraded by neprilysin. So I think it's not intellectually biased to say if you had to choose, it would be NT-proBNP if you were monitoring treatment with sacubitril/valsartan, but again, we don't often have the luxury to choose if our institution is a BNP institution. |
| Robert Christenson (42:20): |
And obviously what Dr. Januzzi is saying, at least my interpretation, is that the BNP half-life is extended. So the BNP's half-life is extended, so it's around a lot longer. NT-proBNP is unaffected, so it falls the way it normally would. So another caveat for having both in your same institution would be if you were taking a ratio between BNP and NT-proBNP, for example, because that ratio would change, would be different in patients that are on ARNIs. |
| Dr. James Januzzi (43:00): |
It's really an interesting topic, right, Rob? |
| Robert Christenson (43:01): |
Yeah. |
| Dr. James Januzzi (43:01): |
You and I both have looked at markers in people treated with this drug, and it's far more complicated than just one goes down, one goes up. And in fact, first of all, the reason for the drop in NT-proBNP is not the increase in BNP. The rise in BNP may not even be significant. We showed in fact that it's probably related to rise of another cyclic GMP stimulator, that being ANP, which rises two to three times higher in people treated with the drug./td>
|
| (43:38): |
But interestingly, other things happen to the natriuretic peptide proteins. First, there's a change in the glycosylation distribution in the end terminal fragment, which may allow for more cleavage and activation of the precursor proBNP(108). In addition, there may be relatively subtle but significant changes in the ratio of circulating BNP and proBNP(108) all driving towards more cyclic GMP production, which to me really indicates that we're still very much in the early days of understanding how neprilysin inhibition may be harnessed to improve outcomes. |
| Chris Driscoll (44:30): |
Thank you. All right, and this question is for both of you. What recent research on NP assays are you excited about from a clinical lab perspective? |
| Robert Christenson (44:41): |
Well, maybe I can start off with that one. I think that there are a number of efforts to develop new assays that perhaps are more predictive. I'm not sure. There were a few that came out some years ago that were championed as being a better mousetrap for monitoring natriuretic peptide release, et cetera. But I don't know that any of them really panned out. Jim, are you more? |
| Dr. James Januzzi (45:19): |
Yeah. Thanks for asking me, Rob. It's hard to improve on a really good working assay, right? And so some of the questions are ... I mentioned glycosylation. To the extent that the amount of sugar that's scattered across the end terminal fragment in particular may interfere with binding of the antibodies used in the immunoassays, there are some who have asked, "Well, does the amount of glycosylation change depending on the physical state of the patient?" |
| (45:55): |
And the answer is, it does actually. And so the very same patient may have rising and falling of the perceived NT-proBNP, for example, because of not less or more production of the peptide, but just more or less glycosylation on the peptide itself. So people have asked, well, maybe we should use antibodies aimed at the non glycosylated portions of NT-proBNP. And there have been assays that have been developed, but they just don't work much differently. |
| (46:24): |
And so at the end of the day, I don't know if it's going to make much of a difference. What I'm really excited about, quite frankly, is number one, ongoing proliferation of outstanding natriuretic peptide assays for both NT-proBNP, and BNP. Multiple vendors are now updating their assays invalidating performance in a contemporary set of patient populations. That to me, is wonderful. |
| (46:49): |
In addition, there's increased interest, as we've already discussed, towards testing populations that had not yet really been evaluated to any great extent, the use in apparently well populations for predicting adverse outcome. There's a lot of information needed to sort of be gained. And then lastly, I'm heavily conflicted here because I'm working on this myself with others. But the extension of natriuretic peptide testing to populations or regions or areas that it just can't reach right now, specifically home-based testing is something that we're really excited about too, because the ability to monitor patients longitudinally in the home rather than bringing them into the lab would be, I think, a real advance for the use of natriuretic peptide testing. |
| Chris Driscoll (47:41): |
Thank you. And I know I do want to get to the Q&A. So folks, while you're listening, feel free to answer, put any questions in the Q&A panel. However, I do want to get one more question in before we turn to our audience, and this is for both of you. From a forward-looking perspective, how do you see heart failure diagnostics and the use of natriuretic peptides evolving 10 years from now, forward-looking? |
| Dr. James Januzzi (48:04): |
Yeah, I said that earlier, if you asked me five years, but we can even say 10. The American Diabetes Association last year recommended yearly NT-proBNP or BNP testing in individuals with chronic diabetes who do not have a diagnosis of heart failure. The American College of Cardiology, American Heart Association, Heart Failure Society of America have made similar recommendations. People at risk, people with diabetes, hypertension, et cetera. |
| (48:34): |
So are we there yet? It's picking up. About 10% of patients with diabetes have had a natriuretic peptide measured in this way in recent TriMedX data. So clinicians are starting to use the test as a screening tool for identifying heart failure at an earlier stage. But there's a lot we need to learn, obviously. What do you do with an abnormal test? How do you respond? For individuals with diabetes we have drugs that actually reduce heart failure risk. So I think the answer's pretty clear. But in a larger sense, if I, without a diagnosis of diabetes, walked into my primary care office and my NT-proBNP was, I don't know, 100, what do you do about that? And that I think is something that clinical trials will have to address. |
| Robert Christenson (49:25): |
I think wellness testing, I think that these NT-proBNP, BNP has been recognized as related to risk and from heart failure being a big cause, but others. So folks that are in their normal state of health that have elevated values, it makes sense that perhaps looking at these individuals early on with early treatment with drugs that are known to reduce heart failure risk or other risk would be prudent. So I'm really excited about that. |
| (50:04): |
Jim, what do you think is the future for actually an algorithm for guiding therapy using ... I know AI and other techniques have been used, and AI, you can't read a sentence without AI mentioned practically these days. So I know you've worked on this a lot over the years, and I don't know what you think the future might be with that. |
| Dr. James Januzzi (50:31): |
Thanks, Rob. And what I will say is that having protocols in place will only serve to improve care for patients. AI is, I think, an ideal option here because it can weigh both the factors leading to higher than expected values combined with the factors that we know are associated with higher levels. In other words, a person comes in with a high value, AI can rapidly evaluate prior measurements, change over time, presence of comorbidities, and interpret the value maybe in a way that may be somewhat more skillful than a clinician who doesn't think about natriuretic peptides very much. |
| (51:14): |
As my research mentee, Dr. Nasrien Ibrahim, once said, she wrote an editorial not long ago where she put a quote that I used one time, "The natriuretic peptide doesn't lie. The blood test isn't rising just out of spite," and so understanding why it's elevated is I think, part and parcel to a good clinician's job. |
| (51:38): |
For guidance of therapy, this issue of intensifying treatment triggered by an elevated natriuretic peptide value, that's something that we explored in chronic heart failure, didn't really take off. But I think what the better way of thinking about natriuretic peptides in terms of monitoring therapy is to identify the lower risk people in whom interventions or other types of evaluations, like serial imaging, stress testing, invasive eval, could be deferred safely. |
| (52:15): |
Whereas people with higher than expected values, rising values over time, those are people in whom intensifying therapy or intensifying monitoring in terms of testing and other strategies to try to stabilize or advance their heart failure care would be indicated. And so in the end, much as the case with many blood tests, because that's really all a biomarker is, enhance clinical decision-making. I think we see natriuretic peptides settling very comfortably in as a front row important test for evaluating and managing heart failure in 2023 and beyond. |
| Robert Christenson (52:57): |
That puts a big responsibility on the lab to make sure we have quality results we can report for the benefit of these patients, so thank you. |
| Dr. James Januzzi (53:07): |
It's been an incredible run, by the way, for bringing cardiology and the lab together, Rob. Wouldn't you say? |
| Robert Christenson (53:13): |
Oh, absolutely. Absolutely. To think 25 years ago, I mean, basically there was a stethoscope for doing this. Remember early on, even from the folks that were heart failure experts, they really didn't think there was much to this biomarker sort of notion. But I think that has played out to be where NT-proBNP, and BNP play an important role. |
English
