Clinical Innovators | Pioneering Novel Cardiac Biomarkers: An Interview with James L. Januzzi, MD
James L. Januzzi, MD, is the Roman W. DeSanctis Endowed Scholar at Massachusetts General Hospital (MGH) and Hutter Family Professor of Medicine at Harvard Medical School, and has been an active clinician at MGH since 2000. Dr. Januzzi has also been team physician for the Boston Red Sox since 2005. Dr. Januzzi is considered a world expert on the topic of biomarker testing in heart disease, with more than 400 publications on the topic. He was a lead author for the recent “Third Universal Definition of Myocardial Infarction,” a section editor for the 2013 ACC/AHA guidelines for heart failure (HF), and is an associate editor for JACC: Heart Failure.
During the course of your career, we have seen tremendous progress in the clinical utility of cardiac biomarkers (particularly with the impact of the natriuretic peptides and troponins on clinical care). What spawned your early interest in cardiac biomarkers?I actually started my research career as a bench researcher. I learned molecular biology techniques and tissue culture skills in the laboratory of Jan Breslow, MD, at the Rockefeller University in New York. Dr. Breslow had a tremendous influence on me, helping to shape my interests in cardiology and specifically research. During residency at Brigham and Women’s Hospital, I continued my bench research experience in the lab of Elazer Edelman, MD. However, all along, I maintained a deep interest in clinical medicine—my father is a highly skilled clinician who I have always greatly admired. Thus, my lifelong exposure to clinical medicine and the rich background in molecular and cellular biology from working with two giants in the field definitely primed my interest in what ultimately became called “translational research”—although that was not a name that was used at that time!
My interest in biomarkers began contemporaneously with the advent of the troponin era. Brigham and Women’s Hospital was an early adopter of clinical troponin testing, and I became fascinated with the potential benefits of this novel, more sensitive, and absolutely cardiac-specific biomarker for the diagnostic evaluation of myocardial infarction. Upon my move to the MGH for fellowship, I immediately became involved in studies of acute coronary syndrome management, and I specifically began focusing on how troponins might be used for therapeutic decision making—an area that we were among the first to explore. Additionally, we were among the first to examine the use of troponin for evaluating risk after cardiac surgery, an area still in need of more in-depth analyses. However, the bigger chapter in my career was soon to follow.
In the early 2000s, I began to follow and study the work of Mark Richards, MD, a pioneer in the field of natriuretic peptide testing in HF. Dr. Richards, who has gone on to be a great friend, colleague, and mentor, was the first to show that measurement of circulating BNP or NT-proBNP could be harnessed to identify or exclude HF. In 2003, we performed the PRIDE study, the first US-based prospective study of NT-proBNP for diagnostic evaluation for HF. Based on the knowledge base that we and others have generated, it seems that prudent use of the natriuretic peptides really does make a difference to clinical care. For this reason, NT-proBNP and BNP have been given Class I support for their use for diagnosis and prognosis in the recent ACC/AHA guidelines for HF, and a Class II for management of HF. Since the beginning, being involved in natriuretic peptide research has been a remarkable ride.
As the number of new biomarkers continues to grow, how can clinicians stay abreast of the trials that may confer the most clinically useful data to incorporate into their practice?
I am concerned that there are far too many new biomarkers, and not enough studies regarding their optimal application. Additionally, the studies out there are variable in quality. Earlier this year, we had a symposium at the Cardiovascular Clinical Trialists meeting to discuss the framework by which biomarker studies should be performed; there is just too much heterogeneity in the design and execution of biomarker studies these days, and quality control for these studies is lacking. This leads to some degree of fatigue for clinicians who may find themselves asking, “Why does this really matter to me?”
My answer would be that clinicians should keep their eye on journals such as JACC, JACC Heart Failure, Circulation, Circulation Heart Failure, the European Heart Journal and the European Journal of Heart Failure. These journals have an extremely high commitment to providing useful and thoughtfully reviewed biomarker data. Additionally, the Cardiosource Biomarker Community (biomarkers.CardioSource.org) is a terrific source of information regarding what’s hot in the biomarker space—it’s not overly esoteric, and is targeted at the practicing clinician. I visit it all the time, and I think it’s great.
You have commented on the challenge of finding well-defined phenotypes for various biomarkers. How do you go about defining an intervention study that targets that biology?I constantly remind my research team that the most crucial goal when looking at a biomarker is to truly understand what a biomarker is “telling us” when we measure it. There’s a lot that goes into this understanding, and it is critical to know what tissues synthesize a particular biomarker and why. For example, BNP and NT-proBNP were originally portrayed as biomarkers of left HF. Sure, that is true, but we now know a broad range of disease states involving essentially every cardiac structure may lead to their release. The good doctor remembers natriuretic peptides are biomarkers of HF; the great doctor remembers they are so much more than that. Just like any diagnostic tool, biomarkers come with a differential diagnosis for their interpretation.
Beyond this, we need to know some fundamental biology: how long does a biomarker circulate, and how important is the degree of change in its day-to-day concentration? Additionally, knowing how we might respond to a worrisome change in the biomarker (using, for example drug therapy changes) is critically important.
All of these pieces of the puzzle must be understood before it is possible to operationalize biomarker measurement to benefit a patient. It’s not straightforward, but it’s an amazing journey to see a marker go from concept to potential reality. Through our 15 years of work in the MGH Heart Center on more than 100 biomarkers, we have gained a good grasp on how to take a novel marker through these steps.
What have you learned during your work at the MGH Heart Center about how to take a novel marker through the necessary steps to make it clinically useful?Most of all, it is about being practical, and answering an existing clinical need. I tell my research group all the time that none of the information above matters if a biomarker tells us something we already know at the bedside—we don’t need a biomarker to be a crutch to replace our judgment. We need biomarkers to supplement that judgment.
A good case in point about how this process is achieved is using NT-proBNP to supplement clinical judgment about titration therapy in HF patients. This is called “guiding” therapy, and is based on the recognition that this very prognostic biomarker also tends to rise and fall with changes in risk, and therapies for HF tend to lower NT-proBNP values.
Conceptually, if a falling NT-proBNP is good, and HF therapies lower NT-proBNP, it stood to reason that using the marker as a ‘barometer’ for therapy success was a testable hypothesis.
Following early studies where we defined the amount of change needed to constitute a significant change in NT-proBNP, as well as the best moment to resample the marker after a therapy change, we proceeded with a proof-of-concept study called PROTECT. This study suggested that NT-proBNP–guided HF care was superior to standard clinical care for taking care of patients with the diagnosis. We are now at the final, pivotal step, performing a study called GUIDE-IT, a trial sponsored by the NHLBI that will examine this question in a large scale, multicenter format.
What is the current focus of your research? What are some of the most promising biomarkers on the horizon?Besides working diligently at the use of NT-proBNP to guide HF therapy in the GUIDE-IT trial, we continue to work to expand the understanding of newer biomarkers that might supplement the natriuretic peptides for testing across a wide range of patient types. Over the years, we have found several compelling markers that may make the translational journey from bench to bedside.
The most interesting new marker that we’ve studied over the years is probably ST2—a biomarker that is linked to the process of tissue fibrosis, and in the heart, ST2 plays a pivotal role in myocardial fibrosis and remodeling. We, and others, have shown ST2 to be wildly prognostic for major adverse cardiovascular events across a wide range of patient types; it is extremely predictive of ventricular remodeling and risk for hospitalization or death in patients with HF, and also presages the onset of cardiac events in normal patients. We have found that ST2 values respond to several HF therapies (most notably beta blockers), and we are now in the planning phases of an intervention study to see if “ST2-guided” HF therapy may improve outcomes.
Other exciting biomarkers include—of course—the newer, more highly sensitive troponin assays. Clinically, these newer troponin methods can measure deeply into the normal range, far lower than previous troponin methods could. We have worked quite extensively with highly sensitive troponin in patients with ischemic heart disease, HF, and even “apparently normal” patients, and find them to provide unique prognostic information.
We have several other markers that we’re examining, including one that appears to predict the presence and severity of diastolic dysfunction, several markers predictive of the onset of renal failure, a new biomarker that appears to prognosticate cardiometabolic outcomes in women, as well as newer markers to predict ventricular remodeling. It’s really quite amazing what we’re seeing. Of course, it remains unclear if any will gain clinical relevance, but that’s the fun part of this job!
My research group, consisting of some of the best and brightest minds here at MGH, is a source of major pride for me, and it’s been an honor and an absolute joy to be involved in this area of research.
What advances in cardiac biomarker testing do you hope to see in the next decade?
I hope to see the concept of biomarker-based “precision medicine” reach widespread clinical acceptance. This will take a great deal of work, but I envision it to be possible, and potentially greatly beneficial to provide a cost-effective and accurate means to improve cardiovascular care in a personalized fashion.
I consider the use of natriuretic peptides to “guide” HF care a necessary first step. I also have confidence that the next generation of novel biomarkers, such as ST2, have great promise to supplement the natriuretic peptides to further improve the accuracy for detecting and addressing deranged cardiovascular biology in a targeted manner. Well-designed studies to address this concept are needed.
Ultimately, I have said the Holy Grail for precision medicine in HF is to use biomarkers to actually prevent the diagnosis. It’s not a stretch—we know that we can predict HF onset using biomarkers in patients who have not a single symptom or sign of the diagnosis. Early returns suggest this goal is achievable: Prof. Ken McDonald in Dublin, Ireland, reported results from a collaborative care study called STOP HF, that used BNP to identify risk for HF in normal subjects; in this trial, BNP use to trigger better care for the at-risk patients significantly reduced onset of HF, a finding that sets the path for even larger trials. The key will be to design these future studies with great attention to detail, adequate powering, and a focus on high-quality execution and clinical relevance.
< Back to Listings