# The Impact of Cocaine Abstinence or Reduced Use on Radiomic Features of Noncalcified Coronary Plaques in HIV-Infected Cocaine Users with Silent Coronary Artery Disease

> **NIH NIH R21** · UNIVERSITY OF MARYLAND BALTIMORE · 2020 · $231,750

## Abstract

ABSTRACT
Despite the success of antiretroviral therapy (ART), individuals living with HIV experience an increased risk of
coronary artery disease (CAD). Although increased arterial inflammation on 18F-FDG-PET was shown to be
associated with high risk plaque features (HRPFs) in those with HIV and subclinical atherosclerosis, there is no
evidence of increased arterial inflammation in optimally treated HIV-infected adults with full viral suppression,
low cardiovascular risk and no known CAD when the imaging results are compared to those from HIV-
uninfected controls. These findings suggest that HIV-associated atherosclerosis may be triggered by factors
other than HIV-associated inflammation such as traditional cardiovascular risk factors or cocaine use. Other
evidence indicates that cocaine is associated with coronary atherosclerosis. It was also shown that cocaine
use modifies the effect of HIV on subclinical coronary atherosclerosis. Supported by DA035632, this research
team enrolled 102 daily cocaine users (including 84 HIV-infected) with contrast-enhanced coronary CT
angiography (CCTA)-confirmed significant (>50% in diameter) coronary stenosis, and employed a cash-
incentive approach to systematically reinforce cocaine abstinence. The study demonstrated that continued
cocaine use led to coronary plaque volume progression, while abstinent or reduced cocaine use led to plaque
volume regression. However, whether cocaine use influences plaque characteristics in HIV-infected cocaine
users has not been investigated. Plaque characteristics are important because rupture-prone coronary plaques
might not necessarily be only those, which cause obstructive stenosis, and that HRPFs are independent of the
severity of coronary artery luminal stenosis. However, identification of the strongest predictor of CAD among
four HRPFs, the napkin-ring sign (NRS), at present, is based only on expert subjective opinion. A landmark
cardiovascular radiomics study calculated 4,440 radiomic features for each coronary plaque and identified a
subset of these parameters which predicted NRS, a qualitative feature of prognostic imaging marker of acute
coronary syndrome. It is quite possible that with the cardiac radiomic approach, the association between
cocaine use and this critical biomarker will be further clarified. The objective of the proposed study is to
examine whether cocaine use influences plaque features in HIV-infected cocaine users. The proposed study
will perform radiomic analyses on the imaging data from 102 participants on DA035632. The specific aims of
this study are: (1) To examine whether continued cocaine use is associated with the NRS and its associated
radiomic features in HIV-infected cocaine users with significant coronary plaques, (2) To examine whether
abstinent or reduced cocaine use ameliorated NRS and its associated radiomic features in HIV-infected
cocaine users with significant coronary plaques, (3) To compare NRS and its associated radiomic features...

## Key facts

- **NIH application ID:** 10148238
- **Project number:** 7R21DA048780-03
- **Recipient organization:** UNIVERSITY OF MARYLAND BALTIMORE
- **Principal Investigator:** SHENGHAN LAI
- **Activity code:** R21 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2020
- **Award amount:** $231,750
- **Award type:** 7
- **Project period:** 2020-05-01 → 2023-03-31

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/10148238

## Citation

> US National Institutes of Health, RePORTER application 10148238, The Impact of Cocaine Abstinence or Reduced Use on Radiomic Features of Noncalcified Coronary Plaques in HIV-Infected Cocaine Users with Silent Coronary Artery Disease (7R21DA048780-03). Retrieved via AI Analytics 2026-05-24 from https://api.ai-analytics.org/grant/nih/10148238. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*