Radiomics Reveals Fuel Behind Coronary Artery Disease Risk: Cocaine

February 16, 2021
Whitney J. Palmer

Using radiomics to assess impact of unconventional risk factors, including cocaine use and HIV infection, potentially introduces a “brave new world” of precision phenotyping in CAD.

Adding radiomics to coronary CT angiography (CTA) shows that risk factors for coronary artery disease (CAD) – particularly unconventional ones like cocaine use and HIV infection – impact atherosclerosis in individual ways.

In a study published Feb. 16 in Radiology, a multi-institutional team of investigators showed that analyzing coronary plaque phenotypes with radiomics could give providers more information about the unique impacts of various CAD risk factors. This knowledge could lead to more personalized treatment – not only for CAD, but potentially for cancer, as well, they said.

Coronary CTA does a good job of revealing how much arterial narrowing a patient has, but that measurement is not always the most effective way to predict whether a patient will have an adverse event like a heart attack, said lead study author Shenghan Lai, M.D., MPH, epidemiology and public health professor at the Institute of Human Virology at the University of Maryland School of Medicine.

Related Content: Use Coronary CTA First for Evaluating Patients with Stable Coronary Artery Disease

“Some people have very bad stenosis where the vessels are 90 percent blocked and do fine, while others with only 40 percent-to-50 percent stenosis die suddenly without warning,” said Lai, who is also an adjunct epidemiology professor at Johns Hopkins Bloomberg School of Public Health. “This [study] shows that not just stenosis, but the nature of the plaque itself, may play a very important role in risk assessment.”

Given that the number of patients dying from CAD is increasing even though it is an avoidable cause of death, the team said, it is critical for providers to get a better understanding of the course of disease. This is the first study to look at using radiomics to expand upon the current imaging biomarkers employed to pinpoint aspects of plaque development that are typically examined at the molecular level.

To pick up on how cocaine use and HIV infection individually contribute to how plaque changes over time, Lai’s team enrolled 300 patients who had CAD confirmed via CT, but their condition was not yet severe enough to actually present symptoms. Of this group, 226 tested positive for HIV, and 174 individuals self-reported using cocaine at least four times a month for six months. Overall, the team reviewed 600 coronary CTAs.

All total, Lai’s team assessed 1,276 radiomic features that were derived from the Heart Study – a longitudinal investigation of HIV and cocaine effects on subclinical CAD. The followed these features for an average of four years.

Using that long-term data, they found that 24 percent of features were associated with at least one risk factor. Specifically, cocaine use was significantly associated with 72 factors – 5.6 percent – but HIV infection was only slightly associated with 1 percent of radiomics features. Still, they said, the study revealed that HIV infection more profoundly affected the CAD present in younger patients – those under age 51.

Related Content: CTA Improves Coronary Disease Risk Assessment

“Cocaine use plays an important role in the pathogenesis of coronary artery disease,” Lai said. “Cocaine users with HIV should abstain from cocaine use to lower the risk of coronary artery disease.”

In addition to these findings, the team’s evaluation revealed four main results:

  • No overlap was detected in radiomic features associated with conventional cardiovascular risk factors or cocaine use or HIV infection, indicating that different risk factors may have distinct contributions to temporal changes in plaque morphologic features.
  • Sex may affect the magnitude of CAD and could change the impact of risk factors, resulting in different phenotypes. This could explain the differences between men and women in adverse cardiovascular outcomes.
  • Risk factor effects can differ throughout life, particularly HIV infection.
  • Interactions between risk factors could be complex, potentially augmenting or diminishing how they affect each other. For example, the team said, the effects of cocaine were only present in individuals with elevated atherosclerotic cardiovascular disease risk.

Based on these results, it is possible, said U. Joseph Schoepf, M.D., director of the cardiovascular imaging division at the Medical University of South Carolina (MUSC), and Tilman Emrich, M.D., a post-doctoral research fellow in cardiovascular imaging at MUSC, that radiomics will pave the way for researchers to identify unknown disease progression pathways, helping to create a more individualized approach to patient care.

“The study…sheds light into a brave new world, where advanced image analysis-based phenotyping may inspire new research avenues in basic science,” they said in an accompanying editorial. “This expands the one-stop shop role of coronary CT angiography from a combined anatomic and functional test to a new field of precision medicine exploring mechanistic insights in disease using radiomic signatures.”

Still, they cautioned, these results should be interpreted carefully. Lai’s study only included African American participants, so the results may not be generalizable to the greater population. In addition, the reproducibility and standardization of the radiomic features must be further researched.

Ultimately, Lai’s team said, its findings indicate that different risk factors likely correspond to specific known or unknown disease progression pathways rather than acting as an interconnected network that contributes to the development of atherosclerosis. This data will help bring a more complete picture of cardiac health into focus.

“We want to figure out why some people die early, why some die suddenly, and why some people go on and one even if they have very significant fixed disease. With radiomics, we can use a CT image or an MR image, because these images have more data than just stenosis,” he said. “The technology is there, that’s not the key obstacle. The key obstacle is that not enough physician-researchers have access to this information.”