ÌÇÐÄvlog

ImportanceÌý Individuals with human immunodeficiency virus (HIV) infection receiving combined antiretroviral therapy (ART) have an increased risk of myocardial infarction. Effects of ART on arterial inflammation among treatment-naive individuals with HIV are unknown.

ObjectiveÌý To determine the effects of newly initiated ART on arterial inflammation and other immune/inflammatory indices in ART-naive patients with HIV infection.

Design, Setting, ParticipantsÌý Twelve treatment-naive HIV-infected individuals underwent fludeoxyglucose F 18 ([¹⁸F]-FDG) positron emission tomographic scanning for assessment of arterial inflammation, coronary computed tomographic angiography for assessment of subclinical atherosclerosis, and systemic immune and metabolic phenotyping before and 6 months after the initiation of therapy with elvitegravir, cobicistat, emtricitabine, and tenofovir disoproxil fumarate (combined ART). Systemic immune and metabolic factors were also assessed in 12 prospectively recruited individuals without HIV serving as controls. The study began July 24, 2012, and was completed May 7, 2015.

InterventionsÌý Combined ART in the HIV-infected cohort.

Main Outcomes and MeasuresÌý The primary outcome was change in aortic target-background ratio (TBR) on [¹⁸F]-FDG-PET with combined ART in the HIV-infected group.

ResultsÌý For the 12 participants with HIV infection (mean (SD) age, 35 [11] years), combined ART suppressed viral load (mean [SD] log viral load, from 4.3 [0.6] to 1.3 [0] copies/mL; P < .001), increased the CD4⁺ T-cell count (median [IQR], from 461 [332-663] to 687 [533-882] cells/mm³; P < .001), and markedly reduced percentages of circulating activated CD4⁺ T cells (human leukocyte antigen-D related [HLA-DR]⁺CD38⁺CD4⁺) (from 3.7 [1.8-5.0] to 1.3 [0.3-2.0]; P = .008) and CD8⁺ T cells (HLA-DR⁺CD38⁺CD8⁺) (from 18.3 [8.1-27.0] to 4.0 [1.5-7.8]; P = .008), increased the percentage of circulating classical CD14⁺CD16^− monocytes (from 85.8 [83.7-90.8] to 91.8 [87.5-93.2]; P = .04), and reduced levels of CXCL10 (mean [SD] log CXCL10, from 2.4 [0.4] to 2.2 [0.4] pg/mL; P = .03). With combined ART, uptake of [¹⁸F]-FDG in the axillary lymph nodes, as measured by TBR, decreased from a median (IQR) of 3.7 (1.3-7.0) at baseline to 1.4 (0.9-1.9; P = .01) at study end. In contrast, no significant decrease was seen in aortic TBR in response to combined ART (mean [SD], 1.9 [0.2]; median [IQR], 2.0 [1.8-2.1] at baseline to 2.2 [0.4]; 2.1 [1.9-2.6], respectively, at study end; P = .04 by 2-way test, P = .98 for test of decrease by 1-way test). Changes in aortic TBR during combined ART were significantly associated with changes in lipoprotein-associated phospholipase A₂ (²Ô = 10; r = 0.67; P = .03). Coronary plaque increased among 3 participants with HIV infection with baseline plaque and developed de novo in 1 participant during combined ART.

Conclusions and RelevanceÌý Newly initiated combined ART in treatment-naive individuals with HIV infection had discordant effects to restore immune function without reducing arterial inflammation. Complementary strategies to reduce arterial inflammation among ART-treated HIV-infected individuals may be needed.

Myocardial infarction rates are increased by 50% in individuals with human immunodeficiency virus (HIV) infection compared with uninfected persons, controlling for traditional cardiovascular disease (CVD) risk factors.^1,2 Mechanisms for increased risk of myocardial infarction in HIV infection remain unclear but may relate in part to the effects of the virus itself and the body’s immune response.^3,4 Effects of antiretroviral therapy (ART) on myocardial infarction risk in HIV are not fully understood.^5,6

We explored the effects of newly initiated integrase-inhibitor–based ART on arterial inflammation by fludeoxyglucose F 18 ([¹⁸F]-FDG) positron emission tomography (PET) among treatment-naive patients with HIV infection without known CVD. Arterial inflammation is a marker of CVD risk in the general population⁷ and is increased among individuals with HIV infection who receive ART.⁸ We hypothesized that initiation of ART in treatment-naive persons would reduce arterial inflammation as well as systemic immune activation/inflammation.

Box Section Ref ID

This study⁹ evaluated the effects of initiation of ART with elvitegravir, cobicistat, emtricitabine, and tenofovir disoproxil fumarate (hereinafter, combined ART) on arterial inflammation by cardiac [¹⁸F]-FDG-PET scanning among ART-naive individuals with HIV infection. The effects of combined ART on coronary plaque observed on coronary computed tomographic angiography (CCTA) and systemic immune and metabolic factors were simultaneously assessed. The study began July 24, 2012, and was completed May 7, 2015. All participants provided written informed consent and received financial compensation. The study was approved by the Massachusetts General Hospital institutional review board.

Twelve ART-naive men with HIV infection initiating combined ART once daily by their treating physician were recruited from infectious disease clinics in Boston. All 12 participants qualified and enrolled at the General Clinical Research Center at Massachusetts General Hospital. In addition, 12 confirmed HIV-negative individuals were recruited for contextualization of data on immune and inflammatory indices in the HIV-infected group (eFigure 1 in the Supplement). Uninfected control participants underwent blood testing only at baseline and did not receive combined ART or undergo [¹⁸F]-FDG-PET or CCTA. Participants in both the treatment and control groups were enrolled based on similar criteria, including age older than 18 years, no current or prior coronary artery disease or significant autoimmune or inflammatory disease, and estimated glomerular filtration rate of 70 mL/min/1.73 m² or more. For HIV-infected participants, assessments were performed at baseline and after 6 months of combined ART. Assessments were delayed beyond 6 months in 3 of 12 participants, 1 of whom did not undergo end-of-study [¹⁸F]-FDG-PET. The median (interquartile range [IQR]) time of follow-up, including these 3 individuals, was 6.4 (6.0-7.4) months.

Individuals with HIV infection underwent [¹⁸F]-FDG-PET scanning⁸ of the aorta, heart, axillary lymph nodes, spleen, and bone marrow. These patients also underwent CCTA¹⁰ (eMethods in the Supplement). Lipid and creatinine levels were determined using standard techniques. The CD4⁺ and CD8⁺ T-cell counts were determined using flow cytometry. The HIV viral load was determined using ultrasensitive reverse transcriptase polymerase chain reaction with a lower limit of detection of 20 copies/mL (Cobas AmpliPrep; Roche Molecular Diagnostics). Flow cytometric analysis of lymphocytes and monocytes was performed, and levels of immune and inflammatory biomarkers were assessed (eMethods in the Supplement).

The prespecified primary end point was change in aortic target-background ratio (TBR) on [¹⁸F]-FDG-PET with combined ART in the HIV-infected group. The study was powered at 80% to detect a decrease in aortic TBR of 0.3 (1-way testing) based on an assumed SD of 0.42.⁸ The study size was chosen a priori to enable detection of a decrease in aortic TBR of 0.3, representing the difference between (ART-treated) HIV-infected and uninfected groups in prior work.⁸ The selected 6-month study duration was long enough to allow for combined ART effects on arterial and systemic inflammation. Other anti-inflammatory strategies have reduced aortic TBR in comparable time frames.¹¹ Two-way tests for change within the HIV group are reported for all variables. A 1-way test for change is also reported for aortic TBR, the primary end point, consistent with the primary study hypothesis. Matched-pairs testing was used to assess change in variables in response to combined ART among ART-naive individuals with HIV infection. A t test was used for log-transformed data; otherwise, Wilcoxon signed rank tests were used. Between-group comparisons of data for the HIV-infection vs noninfection groups were made using a t test for log-transformed data, the Wilcoxon rank sum test as appropriate for continuous data, and the χ² test for categorical data. P < .05 indicates statistical significance. Statistical analyses were performed using SAS JMP software, version 11.0 (SAS Institute Inc).

Participants with HIV infection were ART-naive, with a median (IQR) time since diagnosis of 0.9 (0.2-1.7) years, mean (SD) CD4⁺ count of 483 (166) cells/mm³, and viral load of 4.3 (0.6) log copies. Mean (SD) age was 35 (11) years. The HIV-infected and uninfected groups had very low and similar traditional CVD risk scores (10-year atherosclerotic cardiovascular disease risk score, 2.1% vs 1.8%; P = .91) (eTable 1 in the Supplement). No participants were receiving statin therapy. At baseline, before ART, HIV-infected individuals demonstrated a higher percentage of activated CD4⁺ and CD8⁺ T cells and higher levels of the chemokine CXCL10 compared with controls (Table).

In response to combined ART, aortic TBR increased from a baseline mean (SD) of 1.9 (0.2) (median [IQR], 2.0 [1.8-2.1]) to 2.2 (0.4) (2.1 [1.9-2.6]) (P = .04 using a 2-way test; P = .98 using a 1-way test for decrease) (Figure 1, Figure 2, and Table). A strong association was observed between the increase in aortic TBR and increase in lipoprotein-associated phospholipase A₂ (²Ô = 10; r = 0.67; P = .03). At baseline, 3 of 12 patients (25%) with HIV infection demonstrated subclinical coronary plaque. In addition, 1 HIV-infected participant without baseline plaque developed de novo plaque during the study period. Total, noncalcified, and calcified plaque volumes were higher after therapy among participants with HIV infection who had any plaque at baseline (eTable 2 in the Supplement).

Among participants with HIV infection, there was visible high-level [¹⁸F]-FDG uptake in the bilateral axillary lymph nodes, spleen, and bone marrow. In response to combined ART, the bilateral axillary lymph node TBR was reduced significantly (Figure 1, Figure 2, and Table). There was also a trend toward a reduction in TBR in the spleen but not in the bone marrow (Table).

Treatment with combined ART increased the CD4⁺ count and CD4/CD8 ratio and reduced the viral load (P < .001) as anticipated (Table and eFigure 2 in the Supplement). Viral load, assessed by an ultrasensitive assay, was suppressed to undetectable levels (<20 copies/mL) in 11 of 12 participants and to 25 copies/mL in 1 individual. Combined ART also reduced the percentage of circulating activated CD4⁺ and CD8⁺ T cells as well as the levels of the chemokine CXCL10, albeit not to the levels observed in the control group (Table and eFigure 3 in the Supplement). Finally, combined ART increased the circulating percentage of classical CD14⁺CD16^− monocytes with a trend toward a concomitant decrease in the percentage of CD14⁺CD16⁺ monocytes (Table and eFigure 3 in the Supplement).

With combined ART, high-density lipoprotein cholesterol levels tended to increase and total cholesterol levels increased modestly, but the ratio of total cholesterol to high-density lipoprotein cholesterol did not change. Creatinine levels increased slightly. There was no association between change in metabolic and renal function values and change in aortic TBR. Overall, the 10-year atherosclerotic cardiovascular disease risk score did not change significantly (Table).

Combined ART was well tolerated. No serious adverse events related to combined ART were reported.

Arterial inflammation, reflected in the aortic TBR, did not decrease after treatment with combined ART. In contrast, the TBR of the axillary lymph nodes consistently decreased with combined ART. For context, the increase in aortic TBR among participants initiating combined ART in the present study resulted in a final level of aortic TBR approximately equivalent to that seen among HIV-infected patients receiving long-term ART in a prior study.⁸ We observed a significant correlation between ART-induced changes in 2 separate measures of arterial inflammation–aortic TBR on [¹⁸F]-FDG PET and plasma levels of lipoprotein-associated phospholipase A₂, a marker that relates to incident CVD events in the general population.¹² Although our study was not primarily powered to detect changes in coronary plaque, our CCTA assessments revealed progression of atherosclerotic plaque volume coinciding with ART treatment among individuals with HIV infection who had plaque. Changes in coronary plaque volume occurred over a short time period among young patients with HIV infection who had a low 10-year atherosclerotic cardiovascular disease risk score.

Changes in arterial inflammation and coronary atherosclerosis with combined ART occurred in the context of improved immune homeostasis. In addition to potently suppressing viremia, combined ART increased the CD4⁺ T-cell count and reduced axillary lymph node TBR on [¹⁸F]-FDG-PET. Combined ART also reduced CD4⁺ and CD8⁺ T-cell activation and levels of the chemoxine CXCL10, although not to the levels seen in the controls. Moreover, with therapy, subpopulations of circulating monocytes shifted, favoring a relative increase in the percentage of classical CD14⁺CD16^− monocytes and a decrease in the intermediate/inflammatory CD14⁺CD16⁺ population. For context, previous studies¹³ have demonstrated effects of select ART regimens, including those incorporating the integrase inhibitor raltegravir, to reduce CD4⁺ and CD8⁺ T-cell activation. In previous studies involving other regimens, newly initiated ART did not change the proportions of monocyte subsets but did change patterns of monocyte cell-surface expression.¹⁴

Several possible explanations exist for why arterial inflammation was not reduced in the context of favorable combined ART effects on immune factors. First, combined ART only partially dampened select indices of immune activation. For example, although combined ART reduced T-cell activation, the percentage of circulating activated T cells remained elevated compared with those seen in the control participants. In addition, select monocyte activation markers linked to arterial inflammation⁸ were not reduced by combined ART therapy. Second, effects of integrase inhibitor–based therapy to alter monocyte cell surface receptor expression (eg, expression of CX3CR1, not tested in this study) could influence honing of monocytes to the vascular endothelium.¹⁵ Third, small, anticipated combined ART-mediated effects to increase creatinine levels¹⁶ might be expected to influence arterial inflammation, but in our study, change in creatinine levels did not relate to change in aortic TBR. Finally, it is possible that different results would be observed among patients with a longer duration of HIV infection, longer duration of ART, or higher baseline traditional CVD risk.

Our findings reinforce the overall benefits of immediate ART,¹⁷ including favorable effects of ART to suppress viremia, partially restore immune homeostasis, and partially dampen select immune activation indices. However, our longitudinal study provides novel data derived from [¹⁸F]-FDG-PET and CCTA to suggest for what we believe to be the first time that these effects may be insufficient to forestall the progression of HIV-associated CVD, at least over the short-term. Complementary strategies to further improve immune activation indices and arterial inflammation—including statins and other immunomodulatory strategies—may be needed in addition to ART. Strengths of our study include the novel application of [¹⁸F]-FDG-PET and CCTA in concert with detailed immune and metabolic phenotyping before and after new initiation of a contemporary ART regimen. Limitations of our study include the relatively small sample size and the absence of women in the cohort. Furthermore, we used stand-alone [¹⁸F]-FDG-PET imaging with manual coregistration to CCTA imaging. In addition, the full clinical impact of arterial inflammation as measured by [¹⁸F]-FDG-PET in the HIV population remains unknown.

Studies are needed to examine longer-term, comparative effects of different ART regimens on arterial inflammation. In addition, future studies are needed to elucidate the relationship between arterial inflammation, atherogenesis, and myocardial infarction in HIV.

Corresponding Author: Steven K. Grinspoon, MD, Program in Nutritional Metabolism, Massachusetts General Hospital and Harvard Medical School, 55 Fruit St, 5 Longfellow Pl, Room 207, Boston, MA 02114 (sgrinspoon@partners.org).

Accepted for Publication: March 16, 2016.

Published Online: May 25, 2016. doi:10.1001/jamacardio.2016.0846.

Author Contributions: Drs Tawakol and Grinspoon contributed equally to the study. Dr Grinspoon had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Zanni, Tawakol, Grinspoon.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Zanni, Toribio, Martin, Tawakol, Grinspoon.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Zanni, Toribio, Martin, Lee, Grinspoon.

Obtained funding: Zanni, Melbourne, Tawakol, Grinspoon.

Administrative, technical, or material support: Lu, Martin, Hoffmann.

Study supervision: Grinspoon.

Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Zanni participated in a scientific advisory board meeting for Roche Diagnostics and received grant support from Gilead Sciences, both unrelated to this study. Dr Melbourne is employed by Gilead Sciences. Dr Hoffmann has received grants from HeartFlow Inc, Siemens Healthcare, Genzyme, and the American College of Radiology Imaging Network and personal fees from the American Heart Association, all unrelated to this study. Dr Williams served on the scientific advisory board and was paid by Macrophage Therapeutics LLC, unrelated to the study. Dr Tawakol served as a paid consultant for Actelion, Amgen, AstraZeneca, Cerenis, and Takeda and received grant support from Actelion, Genentech, and Takeda, all unrelated to the study. Dr Grinspoon served as a paid consultant to Gilead Sciences, Theratechnologies, BMS, NovoNordisk, Merck, Navidea, Aileron, and AstraZeneca and received grant support from Amgen, BMS, Gilead Sciences, Kowa Pharmaceuticals, and Theratechnologies, all unrelated to the study. No other disclosures were reported.

Funding/Support: This work was supported by an investigator-initiated grant from Gilead Sciences to Dr Grinspoon and by National Institutes of Health (NIH) grants M01-RR-01066 and 1 UL1 RR025758-01 to the Harvard Clinical and Translational Science Center from the National Center for Research Resources. Dr Zanni was supported by a Medical Research Investigator Training award from the Harvard Catalyst/The Harvard Clinical and Translational Sciences Center (National Center for Research Resources and the National Center for Advancing Translational Sciences, NIH Award 8KL2TR000168-05). Dr Lu was supported by the American Roentgen Ray Society Scholarship.

Role of the Funder/Sponsor: The study funders had no role in the design or conduct of the study; collection, management, analysis, and interpretation of the data; preparation of the manuscript; and decision to submit the manuscript for publication. Gilead Sciences reviewed the manuscript prior to submission, but submission was not contingent on approval by Gilead Sciences.

Disclaimer: The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH or Gilead Sciences.

Additional Contributions: We thank the nursing staff at the Massachusetts General Hospital Clinical Research Center as well as the individuals who participated in this study.