Levels of nine blood proteins can accurately predict high or low risk of serious cardiovascular events
A study published online today in the Journal of the American Medical Association (JAMA) details the first successful effort to measure at scale the levels of more than a thousand proteins in blood in patients with coronary heart disease (CHD), and describes the identification of a combination of nine specific proteins whose levels taken together can accurately predict the risk of future heart attacks, strokes, heart failure or death. This nine-protein model is more accurate for predicting risk of such adverse events than any currently measured risk factors (e.g., cholesterol, blood pressure, diabetes, smoking or age). A diagnostic test based on these findings is in development for release later this year.
“There is a critical need in medicine to correctly identify individuals who are at high risk of devastating cardiovascular events and to be able to tailor treatments to these individuals rather than treating all patients the same,” said Peter Ganz, MD, Chief of the Cardiology Division at Zuckerberg San Francisco General Hospital and Professor of Medicine at University of California, San Francisco, who led the study. “The discovery of different protein levels in blood that can be used to predict the risk of cardiovascular events gives us a much-needed tool to enable more personalized management of individual patients with heart disease.”
The research described in JAMA involved measuring 1,130 different proteins in nearly 2000 individuals with apparently stable coronary heart disease, who were followed up to 11 years. Initially, two hundred different proteins were identified whose blood levels could be related to the risk of heart attacks, strokes, heart failure and death, and ultimately a combination of nine proteins was selected based on their combined accuracy and sensitivity. Application of these findings to CHD patient samples demonstrated that some of those who were deemed clinically stable instead had a high risk of adverse cardiovascular outcomes, while other patients with the same clinical status had a very low risk. Briefly, individuals who all carried the same clinical diagnosis of stable coronary heart disease had a risk of an adverse cardiovascular event that varied by as much as ten-fold, as revealed by analysis of the levels of the nine proteins in their blood.
Repeating this nine-protein measurement over time also showed that people approaching serious cardiovascular events or death had a far greater increase in their protein score than people who were not approaching a new event. The ability to pinpoint those individuals who are at high risk or whose risk is rising may ultimately provide guidance on which patients need to be treated most intensively. It may also guide the appropriate use of medical treatments that are expensive or that carry a significant risk of side effects.
“Existing risk measures based on genomics, clinical risk factors and routine laboratory measures just don’t work very well in patients with coronary heart disease,” said study author Stephen Williams, MD, PhD, Chief Medical Officer of SomaLogic, the company that developed the new protein measurement technology used collaboratively with the UCSF investigators in the JAMA study. “Our approach uniquely allows us to measure thousands of proteins accurately in each patient sample, which provided the basis for identifying the specific proteins that are important for predicting cardiovascular risk.”
This study also marks a significant advance for “personalized” or “precision” medicine. As noted in an accompanying editorial written by Dr. Marc Sabatine of Brigham and Women’s Hospital in Boston, “the quest for personalized medicine is an important one, and the work by Ganz et al. is a welcome step in that direction.” The study’s findings suggest that – at least for some diseases – a protein-based discovery program that does not depend on pre-existing knowledge of pathophysiology can lead to a more personalized risk assessment than any other evaluations available, including those based on genetic findings.
“This has been a particularly gratifying example of a fruitful academic and industry scientific collaboration to pave the way to evaluating true risk across not only cardiovascular disease, but perhaps a wide range of different diseases and conditions,” said Dr. Williams. “We believe that our technology will uniquely accelerate the realization of precision medicine, and are actively engaged with many other academic and corporate partners in several disease areas where there is significant unmet medical need.”
In addition to further evaluation of the results described in the JAMA paper, there is an ongoing discovery program to identify proteins that can predict the risk of cardiovascular disease in additional patient populations, including lower-risk individuals who appear healthy but may actually be at high risk of coronary heart disease due to high cholesterol, high blood pressure, diabetes or smoking, or higher-risk individuals with kidney disease or HIV infection. There are also development efforts underway to make a test based on the findings described in the JAMA paper available through the SomaLogic CLIA laboratory by early fall of this year.
Reference: Ganz P et al. (2016). Development and Validation of a Protein-Based Risk Score for Cardiovascular Outcomes Among Patients with Stable Coronary Disease. Journal of the American Medical Association. Doi:10.1001/jama.2016.5951