A 3D image of a heart is shown.

Accelerating heart failure research

An NIH-supported research partnership aims to transform heart failure research to bring personalized therapies to patients

Every minute of every day the heart pumps about five quarts of blood – nearly 20 glassfuls – throughout the body. But when changes occur, such as the heart becoming weak or not filling adequately with blood, it loses its ability to fully support the body’s needs, causing a condition called heart failure.

Heart failure affects about 6.2 million Americans, including one in five adults older than age 65, and is expected to rise with an aging population. Common symptoms include shortness of breath, fatigue, or swelling in the lower body, such as the legs and feet.

Currently the condition has no cure, so to support earlier detection and treatment, researchers are pooling knowledge across industries and disciplines through a new, potentially game-changing initiative called the Accelerating Medicines Partnership® Heart Failure (AMP® HF) Program.

Supported by the Foundation for the National Institutes of Health, this private-public partnership aims to help researchers identify new, personalized therapies for heart failure – especially for one of the most common types, called heart failure with preserved ejection fraction (HFpEF). HFpEF affects about half of patients with heart failure. However, it can be harder to identify and treat – often because of the complex and different ways it impacts the body.

“The group of patients who have heart failure with preserved ejection fraction is very heterogenous,” said Vandana Sachdev, M.D., a senior research clinician at NHLBI. “The goal is to understand the different subtypes of disease and identify treatment targets for each type.”

Classifying heart failure subtypes

Current heart failure treatment is often based on a person’s cardiac structure and function. To assess function a doctor typically measures the ejection fraction – the percent of blood in the left ventricle that is pumped out of the heart each time it contracts.

If the left ventricle pumps out less than 40% of blood in one beat, a person may be diagnosed with a type of heart failure called heart failure with reduced ejection fraction (HFrEF). People who have coronary heart disease or who have had a heart attack or other conditions that weaken the heart muscle are more likely to have this type.

Heart failure with preserved ejection fraction is different, occurring despite the fact that the heart’s pumping function is normal. HFpEF results when the walls of the heart become too stiff to fully relax between beats, which makes it difficult for the heart to fill up with enough blood.

High blood pressure, obesity, diabetes, and other conditions that contribute to chronic inflammation can stiffen the walls of the heart leading to HFpEF.

Unlike with HFrEF, however, diagnosing and treating HFpEF is tricky. One reason: an ejection fraction of 55% or more is considered normal and a person with HFpEF can appear to have normal heart-pumping function when given cardiac tests.

“The ejection fraction may be normal, but from the patient’s perspective, symptoms of HFpEF and HFrEF are very similar,” Sachdev said. That’s because as blood from the stiff heart backs up into the lungs, people can get short of breath and develop swelling in the legs – which are classic heart failure symptoms. But they can also experience different symptoms, underscoring the challenges of detecting HFpEF.

Studying multiple systems

To categorize HFpEF, researchers are looking for clues from multiple systems that support the body’s functions, including genes, proteins, and gut bacteria. They are then pairing that information with a patient’s physical symptoms, their medical history, and cardiac imaging exams to understand how it affects them.

For example, using cohorts of the Framingham Heart Study, researchers analyzed 71 proteins from more than 7,000 adults, including nearly 200 who later died from heart failure. Through this analysis, they identified five biomarkers associated with HFpEF, which suggested inflammation, stress to the heart, and a hardening of the heart’s muscles and blood vessels were in play. They also found 14 biomarkers associated with HFrEF that provided insight about inflammation and changes in the heart’s structure. Three were linked to both types.

Some of the biomarkers, such as increased levels of natriuretic peptides – hormones secreted from the heart – are used to screen adults for heart failure. The study is now helping explain the role of other biomarkers and how HFpEF differs from HFrEF.

“We still have a lot to learn,” said Jennifer E. Ho, M.D., a cardiologist at Beth Israel Deaconess Medical Center and an associate professor of medicine at Harvard Medical School. Previous research, she said, has found that 87% of these same 71 circulating proteins associated with cardiovascular disease differ between men and women.

However, Ho and others studying biomarkers for heart failure are already finding patterns in these proteins – unrelated to gender – that may be as significant. And they envision that one day physicians could pair these patterns, or “signatures,” with traditional risk factors, like high blood pressure, to better assess a patient’s risk for heart failure.

“In the future, I see that in addition to using a one-size-fits-all approach for the therapies we have available, we’ll also think about other adjunct therapies that we can tailor to individual patients,” Ho said.

Using big data

Through the AMP HF program, results from many previous trials and studies, like those that Ho and others are working on, will be pooled together through a cloud-based system called BioData Catalyst. The digital portal will enable researchers to store, share, and analyze data with others in the program.

“To use new advanced analytic techniques, you need large numbers of patients,” Sachdev said. This is why the HeartShare program, which is the public portion of AMP HF’s public-private-partnership, will enable researchers to analyze data from the electronic health records of 100,000 adults with and without heart failure. Around 10,000 of these adults – those with heart failure – will help researchers analyze HFpEF subtypes.

Within the group of adults with heart failure, 1,000-1,500 with HFpEF will also enroll in a five-year clinical research study. They will undergo comprehensive medical exams, enabling researchers to collect information about their heart, their cardiopulmonary function, and the health of their other organs and systems. Artificial intelligence and machine learning models will be trained to scan echocardiographs and other heart and chest images. These data, along with data from prior studies, will then be added to BioData Catalyst so that scientists can study HFpEF patterns, which they can use to conduct follow-up studies.

Researchers at Northwestern University, one of six study sites, will guide the collection and translation of the HeartShare data, which will be integrated into BioData Catalyst. Massachusetts General Hospital and Brigham and Women’s Hospital, the Mayo Clinic, the University of Pennsylvania, Wake Forest University, and the University of California-Davis will serve as the other study sites.

Built into the HeartShare program, Sachdev said, is the ability to aggregate data not only from multiple sources like medical exams and images, but directly from the trial participants themselves. For example, participants may be able to share information from physical activity monitors through an app called Eureka.

As a first step to enable researchers to access this information, scientists plan to harmonize data, images, and biospecimens from other studies with universal terms and search tools. This will enable researchers to ask questions about HFpEF using the existing datasets, while eliminating the need to create additional background resources for their studies.

Throughout the five-year program, study participants will also be invited to join other trials, such as those that explore connections between HFpEF and sleep disruptions, gut health, genetic and family links, and cognitive dysfunction.

Sachdev hopes that, by the study’s end, researchers will have identified different subtypes of HFpEF and will have learned about treatment targets for individual subtypes so that patients with HFpEF can receive more personalized treatments compared to general medications and devices used today.

Previous studies have started to cluster HFpEF into categories, but more work is needed with diverse populations and to delve deeper into the molecular mechanisms of each subtype.

Sachdev and others envision that bringing researchers from the public and private sectors together may help achieve the biggest goal of all: to create personalized therapeutic options for people living with HFpEF – not just in the United States, but throughout the world.

“We hope the AMP HF partnership will be the stimulus to help us reach that goal,” Sachdev said.


To learn about heart failure, visit https://www.nhlbi.nih.gov/health/heart-failure.

To learn about the AMP HF program, visit https://amphf.org.

ACCELERATING MEDICINES PARTNERSHIP® and AMP® are registered service marks of the U.S. Department of Health and Human Services.


Ask for press officer on duty