CROI 2025 Abstract eBook

Abstract eBook

Poster Abstracts

data were obtained from the Brazilian Census Bureau (IBGE) for the same period The analysis included all causes of death recorded on death certificates (DC), coded according to the International Classification of Diseases, 10th edition (ICD-10) as follows: COVID-19: B342, U071, U072. HIV: B20-B24 Acute Myocardial Infarction (AMI): I20-I25; Stroke: I60-I69; CVD: I00-I99 (excluding I46). We compared two scenarios: one including all deaths and one excluding those in which DC mentioned COVID-19. Mortality ratios (MRs) were calculated relative to 2018. Adjusted MRs were computed using Poisson models with population size as an offset, accounting for state of residence, age group, and sex. Mixed effect models with random intercepts (GLMM) were applied to control for state variations. We also tested the interaction of HIV and year among those without COVID-19. All analyses were conducted with R 4.2.3. Results: A total of 9,063,805 deaths were recorded in Brazil from 2018 through 2023. Among HIV negative individuals (HIVN), the MRs for CVD slightly increased in 2020, peaking in 2021 [1.26 (1.25-1.26)], but remained close to 1, when COVID 19-related deaths were excluded, as was the case of Stroke and AMI. In PLHIV, the scenario differed significantly. The MR for CVD in PLHIV (excluding COVID-19) increased from 1.07 (0.99-1.15) in 2019 to 1.32 (1.24-1.42) in 2023. For CVD, Stroke, and AMI, the annual effect (slope) was significantly different in PLHIV compared to the HIVN (p<0.001), consistently increasing. In contrast, non-HIV deaths either decreased or remained stable when COVID-19 was excluded. Conclusions: Unlike HIVN, PLHIV in Brazil are experiencing a sustained and progressive rise in CVD mortality, independent of COVID-19, that may be linked to broader, underlying factors of worsening cardiovascular health. Effect of Treatment for Opioid Use Disorder on Endothelial Function in People With and Without HIV Jared C. Durieux 1 , Ann K. Avery 2 , Corrilynn O. Hileman 2 1 University Hospitals Cleveland Medical Center, Cleveland, OH, USA, 2 MetroHealth Medical Center, Cleveland, OH, USA Background: Medication-assisted treatment (MAT) with buprenorphine/ naloxone (BUP) or methadone (MET) is an effective harm reduction strategy for people with opioid use disorder. Whether there is a benefit of one over the other on cardiovascular disease (CVD) risk is unknown and is especially important for people with HIV (PWH) where we have shown that use of illicit fentanyl/heroin results in heightened vascular inflammation. Methods: People with and without HIV initiating MAT were enrolled in a 48-week prospective cohort study. Measures of endothelial function (reactive hyperemic index [RHI]) and arterial elasticity (augmentation index standardized at 75 beats per minute [AI] and pulse wave velocity [PWV]) were assessed at entry, week 12, 24, and 48. Constrained longitudinal mixed models with random intercept were used to assess changes in these outcomes. Adjusted models included age at baseline, sex, race, and time-on-MAT. Results: 82 participants (28 MET; 54 BUP) were enrolled. Overall, median (IQR) age was 39 (33, 52) years; 33% were female; 62% were White not Hispanic, 15% were Black not Hispanic, and 21% were Hispanic regardless of race. 21 (26%) participants had HIV; entry current and nadir CD4 counts were 651 and 271 cells/mm 3 and 78% had HIV-1 RNA <200 copies/ml. PWH were more likely to receive BUP. Entry median (IQR) RHI was 2 (1.5, 2.5), AI was 9% (-4, 31), and PWV was 6.8 (6.1, 7.9) m/s without differences between groups. Overall, 33% had abnormal RHI (<=1.67) at entry. In adjusted models, at 48 weeks, the marginal mean difference in RHI decreased from 1.8 to 1.6 [D -0.2 (95% CI: -0.2, 0.6)] among MET treated and nominally increased from 2.09 to 2.1 [D 0.01 (95% CI: -0.3, 0.3)] among BUP treated [adjusted mean difference at 48 weeks MET vs BUP = -0.48 (95% CI: -0.88, -0.08; P=0.03)]. There were no differences in change in AI or PWV over 48 weeks between groups. Conclusions: While measures of arterial stiffness did not differ by MAT used, endothelial function as measured by RHI was better in participants treated with BUP than MET. Medications that bind opioid receptors including MAT have immunomodulatory potential which could alter downstream CVD risk. Buprenorphine based MAT may be a better choice for PWH given heightened CVD risk in this population.

Poster Abstracts

846

Cholesterol Synthesis Pathways in Monocyte-Derived Macrophages Link CMV and HIV to Atherosclerosis Katelyn O'Hare 1 , Brian Richardson 1 , Emily Bowman 2 , Banumathi Tamilselvan 1 , Vargab Baruah 3 , Christine O'Connor 3 , Cheryl Cameron 1 , Mark Cameron 1 , Nicholas Funderburg 2 , Michael L. Freeman 1 1 Case Western Reserve University, Cleveland, OH, USA, 2 Ohio State University, Columbus, OH, USA, 3 Cleveland Clinic Lerner College of Medicine, Cleveland, OH, USA Background: Cytomegalovirus (CMV) is a widespread herpesvirus associated with inflammation and inflammation-related disorders, including cardiovascular disease (CVD). Among people with HIV (PWH), who have a two-fold increase in CVD risk compared to people without HIV (PWoH), latent CMV is nearly universal, and CMV reactivation from latency occurs more frequently, but if and how CMV reactivation contributes to elevated CVD risk in PWH is unclear. Methods: Bulk RNA sequencing was performed on monocyte-derived macrophages (MDMs) that were differentiated from monocytes of CMV seropositive (CMV+) PWoH (n=3M,2F), CMV-seronegative (CMV-) PWoH (n=4M,2F), or CMV+ PWH (n=11M,2F), using autologous serum, a model that preserves the influence of systemic biomarkers found in vivo . Pathway and gene set enrichment analyses were used to compare the MDM transcriptomes with previously published datasets from CVD or in vitro CMV infection. Transcription of CMV genes in MDMs was measured by PCR. Results: When compared to MDMs of CMV- PWoH, we found that over 30% of the differentially upregulated genes in MDMs of CMV+ PWH overlapped with the differentially upregulated genes of MDMs from CMV+ PWoH, indicating a strong influence of CMV in the HIV-associated MDM transcriptome. CMV-related genes were significantly enriched for lipid and cholesterol synthesis pathways and included FDFT1 . In datasets of in vitro CMV infection, FDFT1 expression was associated with productive CMV infection. Despite no clear evidence of viral reactivation in the MDMs, consistent with the low prevalence of latent CMV in circulating monocytes of PWoH, the expression of known CMV activators was enriched while the expression of known CMV repressors was reduced in CMV+ MDMs, suggesting that MDMs from people with CMV are transcriptionally permissive for CMV replication. Finally, a gene set of the top 100 most significantly upregulated genes in MDMs from CMV+ PWoH was enriched in bulk tissue from atherosclerotic plaques compared to peripheral blood cells (NES=1.943, P adj=1.31x10 -3 ) and among macrophages isolated from ruptured atherosclerotic plaques compared to those from stable plaques (NES=1.644, P adj=3.79x10 -5 ), consistent with a proatherogenic role for MDMs in people with CMV. Conclusions: Our findings link an enrichment in cholesterol biosynthesis pathways in MDMs to CMV infection, HIV, and CVD, and suggest that PWH with

845

CROI 2025 257