BACKGROUND Statin therapy lowers the risk of major adverse cardiovascular events (MACE) among people with HIV (PWH). Residual risk pathways contributing to excess MACE beyond LDL-cholesterol (LDL-C) are not well understood. Our objective was to evaluate the association of statin-responsive and other inflammatory and metabolic pathways with MACE in the Randomized Trial to Prevent Vascular Events in HIV (REPRIEVE).METHODS Cox proportional hazards models were used to assess the relationship between MACE and proteomics measurements at study entry and year 2, adjusting for time-updated statin use and the baseline 10-year atherosclerotic cardiovascular disease risk score. We built a machine-learning (ML) model to predict MACE using baseline protein values with significant associations.RESULTS For 765 individuals (age: 50.8 ± 5.9 years, 82% men, 18% women), among 7 proteins changing with statin versus placebo, angiopoietin-related protein 3 (ANGPTL3) related most strongly to MACE (adjusted HR [aHR]: 2.31 per 2-fold-higher levels; 95% CI: 1.11–4.80; P = 0.03), such that lower levels of ANGPTL3 achieved with statin therapy were associated with lower MACE risk. Among 248 proteins that did not change in response to statin therapy, 26 were associated with MACE at a FDR below 0.05. These proteins represented a predominantly humoral immune response, leukocyte chemotaxis, and cytokine pathways. Our proteomics ML model achieved a 10-fold cross-validated concordance index (C-index) of 0.74 ± 0.11 to predict MACE, improving on models using traditional risk prediction scores only (C-index: 0.61 ± 0.18).CONCLUSIONS ANGPTL3, as well as key inflammatory pathways, may contribute to a residual risk of MACE among PWH, beyond LDL-C.TRIAL REGISTRATION ClinicalTrials.gov: NCT02344290.FUNDING NIH, Kowa Pharmaceuticals America, Gilead Sciences, ViiV Healthcare.
Márton Kolossváry, Irini Sereti, Markella V. Zanni, Carl J. Fichtenbaum, Judith A. Aberg, Gerald S. Bloomfield, Carlos D. Malvestutto, Judith S. Currier, Sarah M. Chu, Marissa R. Diggs, Alex B. Lu, Christopher deFilippi, Borek Foldyna, Sara McCallum, Craig A. Sponseller, Michael T. Lu, Pamela S. Douglas, Heather J. Ribaudo, Steven K. Grinspoon
Usage data is cumulative from September 2025 through August 2026.
| Usage | JCI | PMC |
|---|---|---|
| Text version | 3,315 | 153 |
| 991 | 82 | |
| Figure | 607 | 0 |
| Table | 505 | 0 |
| Supplemental data | 1,246 | 3 |
| Citation downloads | 275 | 0 |
| Totals | 6,939 | 238 |
| Total Views | 7,177 | |
Usage information is collected from two different sources: this site (JCI) and Pubmed Central (PMC). JCI information (compiled daily) shows human readership based on methods we employ to screen out robotic usage. PMC information (aggregated monthly) is also similarly screened of robotic usage.
Various methods are used to distinguish robotic usage. For example, Google automatically scans articles to add to its search index and identifies itself as robotic; other services might not clearly identify themselves as robotic, or they are new or unknown as robotic. Because this activity can be misinterpreted as human readership, data may be re-processed periodically to reflect an improved understanding of robotic activity. Because of these factors, readers should consider usage information illustrative but subject to change.