Applying a Broad Multi-omics Approach to Cardiovascular and Pulmonary Diseases

Personalized medicine has been a key aspect of oncology care for over a decade. As research reveals more about the genomics of other conditions, including cardiovascular and pulmonary diseases, understanding the molecular signatures of disease is becoming an increasingly important part of clinical care.

Investigators at Mass General Brigham are at the forefront of many of these efforts. Computational biology and advanced statistical analysis, together with more robust databases, are revealing more about the "omics" of a variety of diseases—including genomics, proteomics, transcriptomics, metabolomics, and more.

"There's a lot of work that goes into linking a specific gene to a specific disease," says physician-scientist Krishna Aragam, MD, MS. "It starts with analyzing the genetic data of large populations to identify DNA variants present more frequently in those with a particular condition than in those without. It then takes some combination of computational and/or functional experiments to confidently tie a gene to the condition of interest. But we believe the effort is well worth it, since our aim is to use this information to better diagnose and treat patients."

Clinical Observations Inform Lab Research—and Vice Versa

Dr. Aragam spends part of his time as a cardiologist at Massachusetts General Hospital seeing patients within the Cardiovascular Disease Prevention Center and the Cardiovascular Genetics Program. He spends the rest of his time leading a research lab situated across the Cardiovascular Research Center at Mass General and the Cardiovascular Disease Initiative at the Broad Institute of MIT and Harvard.

"In the clinic, I often see patients with a family history of cardiovascular disease, and I am trying to tease apart how much of their risk is due to shared lifestyle versus shared genetics," he says. "Seeing and managing these patients definitely informs my research in big-data genomics, where we're looking at hundreds of thousands and sometimes millions of people to figure out which patients are at the highest risk for a disease based on genetic factors."

Patients may come to the Cardiovascular Disease Prevention Center because they have a family history of heart attacks or heart failure, especially at a young age. They may also be referred because they are presenting with early signs of cardiovascular disease (such as high cholesterol) that are out of proportion to their age or lifestyle.

More recently, patients have come to the clinic because genomic testing has revealed a pathogenic variant that could predispose them to heart conditions such as cardiomyopathy. These results may arise as an incidental finding when patients have genetic testing for another issue or may be uncovered through a direct-to-consumer test like 23andMe.

"It's a bit of the Wild West because we're learning so much so quickly about which patients with a 'pathogenic' variant are most likely to manifest the condition," Dr. Aragam says. "Because for most of these genetic changes, it's by no means a 100% guarantee that someone with the genetic change will develop the relevant disease—in fact, it's far from it. Genetic information is one piece of the pie, but we think it's an important piece that, based on the rapidly accruing data, warrants more attention than it's received to date."

Other tests, he adds, include regular EKGs and cardiac imaging.

In the lab, Dr. Aragam is focused on the study of human genetic variation as it relates to cardiovascular disease and risk factors, with a particular emphasis on heart failure and coronary artery disease. A prime interest is in the clinical translation of genetic data to improve risk stratification and management in the context of primary and secondary cardiovascular prevention.

One study he led, published in Nature Genetics, described the discovery and systematic characterization of risk variants and genes for coronary artery disease in nearly 1.4 million participants. That research used datasets from several large genomic studies and biobanks that resulted in more than 210,000 cases of cardiovascular disease.

When it comes to counseling patients on their chances of developing a particular condition, Dr. Aragam is able to provide guidance by applying some of the findings from his population-based research. "This is one situation where polygenic risk scores are becoming relevant," he says. "Developing these scores for a number of diseases and preparing them for clinical deployment is an active effort across Mass General Brigham."

A polygenic risk score, which is based on the total number of DNA-associated changes related to a disease, is one way people can learn about their risk of developing that disease.

Currently, commercial genetic tests are used in Mass General Brigham labs. However, Dr. Aragam and colleagues are developing homegrown polygenic risk scores for conditions such as heart attacks, high blood pressure, high cholesterol, and certain rhythm disturbances. He hopes to make these scores available to Mass General Brigham patients in the very near future.

Developing More Personalized Approaches to Treating COPD

Beyond cardiovascular disease, Brigham and Women's Hospital pulmonologist Matthew Moll, MD, MPH, is focused on developing precision approaches for diagnosing and treating chronic obstructive pulmonary disease (COPD). Doing so would help to identify who is at high risk and eventually lead to a more personalized approach to treating this condition. Dr. Moll co-leads a laboratory that conducts this research with Brigham pulmonologist Michael Hyosang Cho, MD, MPH.

"The burden for this disease is quite high," he says. "One of the big issues is that people progress in different ways, even if they have the same kind of markers for disease severity. In addition, our current one-size-fits-all approach to treating COPD is quite limited because you're not really targeting the underlying biology of the disease."

COPD is driven by genetic and environmental factors, and experts believe that genetics account for about 40% of susceptibility. Only a very small percentage of cases are caused by single monogenic factors. For the rest of the cases, this means models need to consider other biomarkers in addition to gene variants—for example, particular types of inflammatory cells and other transcriptomic signatures that can help predict when a patient will have a decline in lung function.

Dr. Moll and his colleagues have used large datasets to develop polygenic risk scores for COPD, research that was first reported in The Lancet Respiratory Medicine. They found that individuals in the top decile had a eight-fold odds of having COPD. They demonstrated in subsequent publications that the score for measuring COPD risk goes above and beyond family history, that genetics and smoking interact and lead to greater decrements in lung function, and that individuals with higher polygenic risk scores are more likely to develop COPD early in life.

This field of research is leading to repurposing existing drugs for treating COPD as well.

"Most current drugs used to treat COPD focus on treating the symptoms rather than the underlying causes," Dr. Moll says. "We have multiple projects now where we are using advanced network methods to identify drug-repurposing candidates. If we can validate these connections, we can expedite the process of developing new therapies."

For example, the biologic drug dupilumab, which targets interleukin-4 and interleukin-13, was recently found to decrease COPD exacerbations in patients with eosinophil counts higher than 300. The drug may ultimately receive FDA approval for this application. Dr. Moll says these kinds of discoveries can lead not only to future clinical trials but also to off-label uses of established drugs for indications related to COPD.

Dr. Moll credits the Brigham's Channing Division of Network Medicine for enabling much of this research. The division is focused on using an integrated, network-based, systems biology-driven approach to define the etiology of complex diseases, reclassify complex diseases, and develop new treatments and preventive strategies.

"This gives us incredible resources in biobanking and computational biology," he says. "It also allows physicians, PhD scientists, biostatisticians, and talented people with many other backgrounds to form collaborative groups that really enable this kind of research."