We observed higher serum levels of biomarkers of inflammation and immune activation in human immunodeficiency virus (HIV)–infected individuals who reported <100% antiretroviral adherence, even though they were virally suppressed (HIV RNA, <50 copies/mL) at the time of biomarker measurement.
Keywords: adherence, inflammation, antiretroviral therapy
Abstract
Background. Human immunodeficiency virus (HIV)–infected individuals exhibit residual inflammation regardless of virologic suppression. We evaluated whether suboptimal adherence to combination antiretroviral therapy (cART) is associated with greater residual inflammation than optimal adherence, despite virologic suppression.
Methods. Longitudinal self-reported cART adherence data and serum concentrations of 24 biomarkers of inflammation and immune activation were measured at the same study visit in HIV RNA–suppressed (<50 copies/mL) HIV-infected men in the Multicenter AIDS Cohort Study from 1998 to 2009. Associations between dichotomized 6-month (<100% vs 100%) and categorized 4-day (<85%, 85%–99%, and 100%) cART adherence with biomarker concentrations were evaluated.
Results. A total of 912 men provided 2816 person-visits with documented plasma HIV RNA suppression. In adjusted models, person-visits at which <100% cART 6-month adherence was reported had higher concentrations of interleukin 2, 6, and 10, interferon γ, tumor necrosis factor α, and C-reactive protein than person-visits at which 100% cART adherence (P < .05) was reported. These same differences were observed in person-visits reporting <85% versus 100% 4-day cART adherence, but not in visits reporting 85%–99% versus 100% cART adherence. After adjustment for multiple comparisons, tumor necrosis factor α remained significantly higher (11% increase; P < .001) in person-visits at which <100% adherence was reported.
Conclusions. Higher concentrations of inflammatory biomarkers were observed among HIV RNA–suppressed men who reported <100% cART adherence than among more adherent men. Residual HIV replication (ie, below the limit of detection), more likely among men with suboptimal adherence, is a plausible mechanism. Whether improving cART adherence could affect residual inflammation and associated morbidity and mortality rates should be investigated.
Sustained use of effective combination antiretroviral therapy (cART) is essential to achieving maximal efficacy in human immunodeficiency virus (HIV) treatment [1, 2]. Systemic exposure to antiretroviral therapy is directly related to host factors including age, sex, weight, diet, genetics, and drug-drug interactions; however, the dominant factor affecting long-term drug exposure is adherence [3]. Although modern cART regimens are more forgiving of suboptimal (ie, less than daily) drug intake, adherence remains the main predictor of HIV outcomes among cART-treated persons [3, 4]. Despite this, little is known regarding non-AIDS biological and clinical consequences associated with variations in adherence.
Initiation of cART and achievement of viral suppression has been associated with reductions in systemic inflammation and immune activation in HIV-infected individuals [5–7]. However, cART-induced viral suppression does not reduce inflammation to levels observed in HIV-uninfected persons, even in the setting of sustained viral suppression [7–10]. The state of persistent inflammation and immune activation has been linked to the development of non-AIDS adverse events including cardiovascular disease, end-stage renal disease, cognitive decline, frailty, and cancer [9, 11–14]. Although the mechanisms behind this are not fully understood, low-level HIV viremia (viral replication below the limits of detection of most commercially available assays) may contribute to this phenomenon by inducing intermittent bursts of inflammation and immune activation [15, 16].
The mechanisms of low-level viremia in chronically virally suppressed, HIV-infected individuals receiving cART also remain poorly understood, although variations in cART adherence could be a contributing factor [17–20]. Thus, it is feasible that suboptimal cART adherence, even among virologically suppressed individuals, could lead to persistently higher levels of inflammation and immune activation, and thereby increased risk for non-AIDS adverse events. Clear links between suboptimal adherence to cART and immune activation and/or inflammation among virologically suppressed HIV-infected individuals are lacking. In the current study, we investigated whether variations in cART adherence are associated with greater levels of inflammation and immune activation among HIV-infected men, independent of plasma HIV RNA suppression.
METHODS
Study Design and Participants
We evaluated prospectively collected longitudinal self-reported cART adherence data from men enrolled in the Multicenter AIDS Cohort Study (MACS; http://aidscohortstudy.org/) between October 1998 and September 2009. The MACS is an ongoing study of HIV-1 infection among men who have sex with men, with 4 US sites: Baltimore, MD/Washington, DC; Chicago, Illinois; Los Angeles, California; and Pittsburgh, Pennsylvania [21]. Briefly, participants are evaluated at 6-month intervals, and study visits include standardized interviews, physical examinations, and blood collection for concurrent laboratory analyses and storage. The MACS protocols were approved by the local institutional review boards at each study site, and informed consent was obtained from all participants before enrollment.
Our study population was restricted to HIV-infected men who reported taking cART and whose plasma HIV RNA levels were <50 copies/mL (Roche Amplicor assay) at the time of their study visit. These person-visits were further restricted to men with available biomarker measurements from a previous MACS study of inflammation and immune activation [7]. cART was defined as a regimen that contained ≥3 antiretroviral drugs, including 2 nucleoside reverse-transcriptase inhibitors with either an unboosted protease inhibitor (PI), a boosted PI, or a nonnucleoside reverse-transcriptase inhibitor (NNRTI), or a nucleoside reverse-transcriptase inhibitor–only regimen. Because of the time period evaluated, this analysis did not include men taking integrase strand transfer inhibitors (INSTIs).
Antiretroviral Adherence Evaluation
Adherence to cART was measured using self-reported data collected at each study visit. Men were asked about the number of pills taken over the prior 4 days for each medication in their cART regimen and whether their cART usage in the 4 days was typical of their use since their prior study visit 6 months earlier. Two measures of adherence were calculated: a dichotomous 6-month adherence variable (based on whether their 4-day cART intake was typical since the last study visit) and a categorical 4-day adherence variable (based on a percentage of the number of pills taken vs prescribed in the last 4 days). For the dichotomous 6-month adherence variable, men were classified as 100% adherent if they reported no missed doses in the past 4 days and also reported that this pattern was typical of the time since last study visit. Men with any other response were assigned <100% adherence. For the categorical 4-day adherence variable, a percentage of expected adherence was calculated as described previously in the MACS: (∑ No. of times drug taken over 4 days)/(∑ No. of times per day drug prescribed · 4) · 100 [4]. When different values of adherence were reported across drugs, the lowest adherence percentage was used. We then classified the percentage into 3 groups based on data suggesting that these are clinically significant thresholds: 100%, 85%–99%, and <85% [3, 4, 22–24].
Biomarkers of Inflammation and Immune Activation
Serologic markers of inflammation and immune activation were quantified using the MesoScale Discovery and Luminex (Luminex) platforms, as described elsewhere [7]. Serum levels of interleukin 1β (IL-1β), interleukin 2 (IL-2), interleukin 6 (IL-6), interleukin 10 (IL-10), interleukin 12p70 (IL-12p70), tumor necrosis factor (TNF) α, granulocyte-macrophage colony-stimulating factor, and interferon (IFN) γ were measured using the Human Pro Inflammatory 9-Plex Ultra-Sensitive Kit (MesoScale Discovery). Chemokine (CC motif) ligand (CCL) 2, CCL4, CCL11, CCL13, CCL17, chemokine (CXC motif) ligand 10 (CXCL10), and interleukin 8 were measured using the Human Chemokine 7-Plex Ultra-Sensitive Kit (MesoScale Discovery). Soluble CD14, CD27, glycoprotein 130, IL-2 receptor α, IL-6 receptor, and TNF receptor 2; B-cell activating factor; and CXCL13 were measured using the Luminex platform. To minimize variability, all samples from a participant were tested on a single plate. Levels of C-reactive protein (CRP) were measured using a high-sensitivity immunonephelometric assay performed through a clinical reference laboratory (Quest Diagnostics).
Statistical Analysis and Covariate Definitions
We included possible confounding covariates based on examination of covariate-adherence and covariate-biomarker relationships and describe only those included in the final analysis. Age was treated as continuous. Race was defined by self-report and was dichotomized as white or nonwhite. Infection with hepatitis C virus (HCV) was defined by the presence of detectable plasma HCV RNA. Tobacco smoking at time of visit was based on self-report and treated as dichotomous (yes or no). Depressive symptoms were defined as a Center for Epidemiologic Studies Depression Scale score ≥16; diabetes mellitus, as a hemoglobin A1C level ≥6.5%, a fasting glucose level ≥126 mg/dL, or the use of antidiabetic medications; anemia, as a hemoglobin concentration below the 5th percentile for the general population; and hypertension, as systolic blood pressure ≥140 mm Hg, diastolic blood pressure ≥90 mm Hg, or the use of antihypertensive medications. Absolute CD4+ T-lymphocyte counts were determined by means of flow cytometry and classified as >500, 351–500, 201–350, or ≤200 cells/μL.
Biomarker concentrations exhibited heterogeneous distributions that were not always log normal. We therefore modeled these concentrations as generalized gamma, a flexible 3-parameter distribution encompassing the log-normal, Weibull, and exponential distributions [25], to avoid imposing strong distributional assumptions on the data. In multivariate models, covariates modified the location (β) parameter of each biomarker distribution. With scale (σ) and shape (λ) parameters held constant, the effect of a covariate on the β parameter may be interpreted as a constant percentage shift in a biomarker distribution across percentiles of that distribution. Biomarker values that were below the lower limit of detection for a given assay were handled by modeling the inverse of concentrations and thus using standard methods for right-censored survival data.
We adjusted models for multiple measurements per individual. Because we tested relationships between adherence and 24 different biomarker concentrations, we adjusted for multiple tests by controlling the false discovery rate at 5%, using the Benjamini-Hochberg procedure [26]. Analyses were conducted using SAS v9.4 (SAS Institute) and Stata 13 (StataCorp) software.
Sensitivity Analyses
We performed supplementary analyses to examine the robustness of our findings to alternate assumptions. We tested whether statin use may have confounded any associations between adherence and biomarker concentrations. We also tested whether estimated effects of lower adherence on biomarker concentrations differed by type of cART regimen, restricting regimens to either ritonavir-boosted PIs or NNRTIs, and by time receiving therapy. We also strengthened our definition of HIV suppressed visits by restricting the person-visits to those from men with (1) undetectable plasma HIV RNA, (2) no prior visit with detectable HIV RNA (during cART), and (3) no time gaps >1 year without an HIV RNA measurement, all within the previous 5 years.
RESULTS
Study Population
We analyzed data derived from 912 men who contributed a total of 2816 person-visits from 1998 to 2009 (median year, 2006; interquartile range [IQR], 2003–2008) at which HIV viral suppression was documented. Participant demographics and person-visit characteristics are shown in Table 1. Each participant contributed a median (IQR) of 3 (2–4) visits, and had accrued a median of 5.4 (2.9–8.0) years of cART use at the time of each visit. The median (IQR) age across visits was 48.4 (42.6–54.0) years, and the median CD4+ T-lymphocyte count was 584 (425–775) cells/μL. Imperfect (<100%) and 100% 6-month adherence were reported in 362 (13%) and 2454 (87%) person-visits, respectively. Based on the 4-day adherence only, 100% adherence was reported in 2491 person-visits (88%), 85%–99% adherence in 112 (4%), and <85% adherence in 213 (8%). Discordant adherence across antiretroviral drug types was reported in 180 person-visits (6%) (lowest level used). Distributions of biomarker concentrations are reported in Supplementary Table 1. Eight biomarkers had ≥1% of measurements below the lower limit of detection at the time of study visit. The proportions of such measurements for these biomarkers were as follows: IL-1β, 43%; IFN-γ, 42%; granulocyte-macrophage colony-stimulating factor, 38%; IL-2, 28%; IL-12p70, 13%; CRP, 4%; IL-10, 2%; and IL-6, 1%.
Table 1.
Characteristics of Study Population and Person-Visits
| Characteristic | Participants or Person-Visits, No (%) |
|---|---|
| Unique MACS participants (n = 912) | |
| White, non-Hispanic | 532 (58) |
| Black, non-Hispanic | 221 (24) |
| Hispanic | 143 (16) |
| Other race | 16 (2) |
| Person-visits (n = 2816) | |
| cART at visit | |
| PI based, boosted | 965 (34) |
| PI based, not boosted | 437 (16) |
| NNRTI-based without PI | 1318 (47) |
| NRTI/other cART without PI | 96 (3) |
| Other factors at visita | |
| HCV infection | 215 (8) |
| HBV infection | 127 (5) |
| Depressive symptoms | 684 (24) |
| Tobacco smoking | 790 (28) |
| Obesity | 310 (11) |
| Diabetes mellitus | 304 (11) |
| Anemia | 362 (13) |
| Hypertension | 623 (22) |
| Statin use | 826 (29) |
| CD4+ T-lymphocyte count at visit | |
| >500/µL | 1767 (63) |
| 351–500/µL | 619 (22) |
| 201–350/µL | 319 (11) |
| ≤200/µL | 111 (4) |
| 100% 6-mo cART adherence at visit | 2454 (87) |
| 4-d cART adherence at visit | |
| 100% | 2491 (88) |
| 85%–99% | 112 (4) |
| <85% | 213 (8) |
Abbreviations: cART, combination antiretroviral therapy; HBV, hepatitis B virus; HCV, hepatitis C virus; MACS, Multicenter AIDS Cohort Study; NNRTI, nonnucleoside reverse-transcriptase inhibitor; NRTI, nucleoside reverse-transcriptase inhibitor; PI, protease inhibitor.
a Depressive symptoms were defined as a Center for Epidemiologic Studies Depression Scale score ≥16; obesity, as a body mass index >30 kg/m2; diabetes mellitus, as a hemoglobin A1C level ≥6.5%, a fasting glucose level ≥126 mg/dL, or the use of antidiabetic medications; anemia, as a hemoglobin level below the 5th percentile for the general population; and hypertension, as systolic blood pressure ≥140 mm Hg, diastolic blood pressure ≥90 mm Hg, or use of antihypertensive medications.
Relationships Between cART Adherence and Biomarker Concentrations
Models estimating biomarker concentrations as a function of 6-month adherence were adjusted for age, race, HCV infection, smoking, depressive symptoms, diabetes mellitus, anemia, hypertension, and CD4+ cell count. Imperfect adherence was associated with higher concentrations of 21 of the 24 biomarkers and significantly associated with higher concentrations of CRP (21% increase; P = .006), IFN-γ (15%; P = .008), IL-2 (14%; P = .02), IL-6 (12%; P = .01), TNF-α (11%; P < .001), and IL-10 (11%; P = .02), relative to 100% adherence (Figure 1, Supplementary Table 2). After adjustment for multiple comparisons, <100% adherence was significantly associated with higher concentrations of TNF-α.
Figure 1.
Percentage shifts in distribution of biomarker concentrations associated with <100% 6-month adherence to combination antiretroviral therapy (cART), compared with 100% adherence. Biomarker data were analyzed at person-visits where human immunodeficiency virus (HIV)–infected men reported taking cART and had plasma HIV RNA levels <50 copies/mL. Generalized gamma models were adjusted for age, race, hepatitis C virus infection, smoking, depressive symptoms, diabetes mellitus, anemia, hypertension, and CD4+ T-lymphocyte cell count. Error bars represent 95% confidence intervals; orange squares, hazard ratios that are statistically significant (P <.05); and red square, hazard ratio that is statistically significant after adjustment for multiple tests, using the Benjamini-Hochberg procedure to control the false discovery rate at 5% [26]. Abbreviations: BAFF, B-cell activating factor; CCL, chemokine CC motif ligand; CRP, C-reactive protein; CXCL, chemokine CXC motif ligand; GM-CSF, granulocyte-macrophage colony-stimulating factor; IFN, interferon; IL-1β, interleukin 1β; IL-2, interleukin 2; IL-6, interleukin 6; IL-8, interleukin 8; IL-10, interleukin 10; IL-12p70, interleukin 12p70; sCD14, soluble CD14; sCD27, soluble CD27; sGP130, soluble glycoprotein 130; sIL-2Rα, soluble IL-2 receptor α; sIL-6R, soluble IL-6 receptor; sTNF-R2, soluble tumor necrosis factor receptor 2; TNF-α, tumor necrosis factor α.
The estimated effects on biomarker concentrations associated with the categorical 4-day adherence from the multivariate models are displayed in Figure 2 and Supplementary Table 3. Adherence between 85% and 99% was not significantly associated with concentrations of any biomarker relative to 100% adherence, with the exception of TNF-α (10% increase; P = .02). By contrast, adherence <85% was significantly associated with higher concentrations of 6 biomarkers relative to 100% adherence (same biomarkers found in the 6-month analysis): CRP (22% increase; P = .02), IL-2 (20%; P = .01), IFN-γ (17%; P = .01), IL-6 (16%; P = .01), IL-10 (13%; P = .04), and TNF-α (10%; P = .001). As in the 6-month analysis, the estimate for TNF-α remained significant after adjustment for multiple comparisons.
Figure 2.
Percentage shifts in distribution of biomarker concentrations associated with 85%–99% and <85% 4-day adherence to combination antiretroviral therapy (cART), compared with 100% adherence. Biomarker data were analyzed at person-visits where human immunodeficiency virus (HIV)–infected men reported taking cART and had plasma HIV RNA levels <50 copies/mL. Generalized gamma models were adjusted for age, race, hepatitis C virus infection, smoking, depressive symptoms, diabetes mellitus, anemia, hypertension, and CD4+ T-lymphocyte cell count. Error bars represent 95% confidence intervals; squares and triangles, <85% and 85%–99% 4-day adherence, respectively; orange symbols, hazard ratios that are statistically significant (P <.05); and red symbol, hazard ratio that is statistically significant after adjustment for multiple tests, using the Benjamini-Hochberg procedure to control the false discovery rate at 5% [26]. Abbreviations: BAFF, B-cell activating factor; CCL, chemokine CC motif ligand; CRP, C-reactive protein CXCL, chemokine CXC motif ligand; GM-CSF, granulocyte-macrophage colony-stimulating factor; IFN, interferon; IL-1β, interleukin 1β; IL-2, interleukin 2; IL-6, interleukin 6; IL-8, interleukin 8; IL-10, interleukin 10; IL-12p70, interleukin 12p70; sCD14, soluble CD14; sCD27, soluble CD27; sGP130, soluble glycoprotein 130; sIL-2Rα, soluble IL-2 receptor α; sIL-6R, soluble IL-6 receptor; sTNF-R2, soluble tumor necrosis factor receptor 2; TNF-α, tumor necrosis factor α.
Sensitivity Analyses
We also adjusted models for statin use, which reduced the person-visits by 256 (owing to missing data). However, our findings were nearly identical and inferences remained unchanged (data not shown). We also restricted the study population to persistently virologically suppressed person-visits; among this group (n = 1279 person-visits), associations between suboptimal 6-month adherence and the previously identified biomarkers were similar, if not stronger (Supplementary Figure 1; Supplementary Table 4). TNF-α concentrations were 15% higher in person-visits reporting <100% adherence (P < .001), a finding that remained significant after adjustment for multiple comparisons. In addition, imperfect 6-month adherence was significantly associated with higher concentrations of IL-2 (26% increase; P = .007), CRP (25%; P = .03), IFN-γ (22%; P = .008), and IL-6 (18%; P = .02), plus 3 additional biomarkers, sCD27 (7%; P = .03), CXCL13 (6%; P = .03), and soluble IL-2 receptor α (6%; P = .03).
We also restricted the person-visits to those in which participants were receiving cART that included either a ritonavir-boosted PI and/or an NNRTI-based regimen (n = 2283) and evaluated whether the type of cART regimen altered the effect of lower adherence on biomarker levels (Supplementary Table 5). On the multiplicative scale, there was evidence for a significant interaction between cART regimen and nonadherence for only IFN-γ and CCL4. In both cases, lower adherence among men taking NNRTI-based cART was associated with greater increases in biomarker concentrations than in men taking a ritonavir-boosted PI regimen (IFN-γ, 30% higher [P = .03]; CCL4, 22% higher [P = .009]). Finally, the addition of time receiving therapy to our model did not affect the point estimates for adherence, and inferences remained identical (data not shown).
DISCUSSION
In this study, we identified a positive association between suboptimal cART adherence and higher levels of inflammation among HIV RNA–suppressed, HIV-infected men receiving cART. We initially found that <100% adherence was associated with higher levels of TNF-α, IFN-γ, CRP, IL-2, IL-6, and IL-10. Further analysis revealed that these associations were largely driven by adherence levels <85%. These associations remained significant after adjustment for various potential confounding factors that can be associated with increased inflammation and, for TNF-α, even after adjustment for multiple comparisons. In addition, these findings were unchanged after controlling for statin use, which can exert an anti-inflammatory effect [27], and after restricting the analysis to persistently HIV RNA–suppressed individuals. To our knowledge, this is the first report in which suboptimal cART adherence has been associated with heightened levels of inflammation and immune activation despite suppressed HIV viremia using standard clinical assays.
Our findings suggest that cART adherence variations could have significant biological consequences despite apparent HIV suppression, because persistent inflammation and immune activation are associated with increased morbidity and mortality among HIV-infected persons [9, 10, 28]. Although the mechanism behind this association remains unclear, residual and/or intermittent (ie, unmeasured) viral replication below the threshold of detection of conventional assays is a likely explanation that should be evaluated. Recent data from another cohort have shown that suboptimal cART adherence is associated with residual plasma viremia (quantified by ultrasensitive HIV RNA single-copy assays), despite apparent virologic suppression and regardless of cART regimen type [18]. Residual viremia has been associated with residual inflammation, increased intestinal microbial translocation, and increased cardiovascular morbidity [29, 30]. Thus, it is plausible that suboptimal cART adherence could lead to episodes of low-level HIV replication and consequent enhanced inflammation and immune activation among HIV-infected individuals who seem to remain virologically suppressed during cART. Although our data alone cannot prove this mechanism, recent findings in a small group of HIV virologic “elite” controllers (HIV load, <40 copies/mL), in whom cART initiation decreased immune activation, are consistent with this hypothesis [31].
To date, HIV viral suppression (by clinically available assays) has been used as the primary surrogate clinical marker of cART adherence, and maintenance of undetectable HIV RNA has been presumed to indicate a level of cART adherence that is “sufficient” to avoid the adverse effects of viral replication, even if an individual is not 100% adherent [23]. Various studies have demonstrated that the level of cART adherence required to maintain virologic suppression decreases with the duration of cART-induced virologic control [32, 33]. In this context, the association between suboptimal adherence and increased inflammation could have unique clinical significance, because declining cART adherence could contribute to levels of residual inflammation observed among HIV-infected individuals despite long-standing apparent viral suppression. It could also explain why reductions in inflammation to levels observed among HIV-uninfected persons have proved difficult to achieve among cART-treated HIV-infected persons [34, 35]. Thus, our data support routine consideration of levels of cART adherence in current and future studies evaluating HIV-associated chronic inflammation.
The biomarkers associated with cART adherence in this study reflect a wide range of inflammatory and clinical pathways that could be heightened in the nonadherent population. For example, higher IFN-γ levels might indicate an ongoing endogenous anti-HIV response (perhaps driven by persistent HIV replication), whereas higher TNF-γ levels suggest activation of innate and adaptive immunity [36]. Similarly, elevated IL-6 and CRP levels suggest responses to an ongoing stimulus (eg, residual HIV viremia) that could result in endothelial inflammation and atherosclerosis [37]. Collectively, these inflammatory and immune activation responses could contribute to the high rate of noninfectious outcomes and clinical outcomes, not typically associated with HIV infection observed in excess (compared with HIV-uninfected persons) among HIV-infected individuals who seem to remain virologically suppressed.
Of particular interest in our study was the finding that increases in inflammatory biomarkers persisted after we restricted our analysis to individuals with long-term HIV suppression. This suggests that the negative effects of suboptimal adherence on residual inflammation and immune activation include a patient population generally presumed to have the highest level of cART adherence, and it supports the premise that incomplete adherence, although sufficient to achieve and sustain viral suppression as shown by conventional assays, may have significant detrimental consequences not previously identified. Whether adherence intensification in virologically suppressed persons receiving cART could translate into a decrease in chronic inflammation remains unclear but warrants future study.
The strengths of our study include its prospective nature in addition to the large sample size and the comprehensive inflammatory biomarker profile obtained. Among its main limitations is the fact that self-report may overestimate cART adherence [38, 39]; however, any such misclassification may have attenuated effect estimates relative to those that could be obtained if better measures of adherence were available. In addition, we did not include individuals taking INSTIs in this analysis. Recent reports have demonstrated that INSTIs could have a greater effect on residual inflammation than NNRTIs owing to their potency and more forgiving pharmacokinetics [6]. Thus, the evaluation of these findings in individuals receiving long-term INSTI-based cART is needed. Furthermore, our study did not evaluate whether low levels of cART adherence were associated with the development of non-AIDS clinical outcomes, which is worthy of further investigation. Finally, residual potential confounding associated with low adherence and high levels of inflammation, such as poor overall health status, concomitant risk factors or lifestyle (ie, exercise, diet), could not be evaluated systematically in this population.
In summary, we demonstrated that suboptimal cART adherence is associated with enhanced inflammation and immune activation despite apparent HIV virologic suppression. Our findings set the framework to better explain the biological consequences of cART adherence variations and have identified adherence as a target for future interventions aimed at further reducing residual chronic inflammation and immune activation in HIV-infected individuals.
Supplementary Data
Supplementary materials are available at http://cid.oxfordjournals.org. Consisting of data provided by the author to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the author, so questions or comments should be addressed to the author.
Notes
Acknowledgments. Data in this manuscript were collected by the Multicenter AIDS Cohort Study (MACS); investigators are listed here by center. Baltimore (U01-AI35042), Johns Hopkins University Bloomberg School of Public Health: Joseph B. Margolick (principal investigator), Jay Bream, Todd Brown, Barbara Crain, Adrian Dobs, Michelle Estrella, W. David Hardy, Lisette Johnson-Hill, Sean Leng, Anne Monroe, Cynthia Munro, Michael W. Plankey, Wendy Post, Ned Sacktor, Jennifer Schrack, and Chloe Thio. Chicago (U01-AI35039), Feinberg School of Medicine, Northwestern University, and Cook County Bureau of Health Services: Steven M. Wolinsky (principal investigator), John P. Phair, Sheila Badri, Dana Gabuzda, Frank J. Palella Jr, Sudhir Penugonda, Susheel Reddy, Matthew Stephens, and Linda Teplin. Los Angeles (U01-AI35040), University of California, UCLA Schools of Public Health and Medicine: Roger Detels (principal investigator), Otoniel Martínez-Maza (co–principal investigator), Aaron Aronow, Peter Anton, Robert Bolan, Elizabeth Breen, Anthony Butch, Shehnaz Hussain, Beth Jamieson, Eric N. Miller, John Oishi, Harry Vinters, Dorothy Wiley, Mallory Witt, Otto Yang, Stephen Young, and Zuo Feng Zhang. Pittsburgh (U01-AI35041), University of Pittsburgh, Graduate School of Public Health: Charles R. Rinaldo (principal investigator), Lawrence A. Kingsley (co–principal investigator), James T. Becker, Phalguni Gupta, Kenneth Ho, Susan Koletar, Jeremy J. Martinson, John W. Mellors, Anthony J. Silvestre, and Ronald D. Stall. Data coordinating center (UM1-AI35043), Johns Hopkins University Bloomberg School of Public Health: Lisa P. Jacobson (principal investigator), Gypsyamber D'Souza (co–principal investigator), Alison, Abraham, Keri Althoff, Jennifer Deal, Priya Duggal, Sabina Haberlen, Eithne Keelagan, Alvaro Muñoz, Derek Ng, Eric C. Seaberg, Sol Su, and Pamela Surkan; National Institute of Allergy and Infectious Diseases (NIAID): Robin E. Huebner; and National Cancer Institute: Geraldina Dominguez.
Disclaimer. The contents of this publication are solely the responsibility of the authors and do not represent the official views of the National Institutes of Health (NIH), Johns Hopkins Institute for Clinical and Translational Research, or the National Center for Advancing Translational Sciences.
Financial support. The MACS is funded primarily by NIAID, with additional cofunding from the National Cancer Institute, the National Institute on Drug Abuse (NIDA), and the National Institute of Mental Health (NIMH). Targeted supplemental funding for specific projects was also provided by the National Heart, Lung, and Blood Institute and the National Institute on Deafness and Communication Disorders. MACS data collection is also supported by the National Center for Advancing Translational Sciences, a component of the NIH (grant UL1-TR001079 to Johns Hopkins University Institute for Clinical and Translational Research), and the NIH Roadmap for Medical Research. The research was also supported by the HIV Prevention Trials Network, sponsored by the NIAID, the NIDA, the NIMH, and the Office of AIDS Research, NIH (grant UM1-AI068613) and by the NIAID (grants K23 AI104315 to J. R. C.-M. and K24 AI120834 to T. T. B.).
Potential conflicts of interest. T. T. B. has served as a consultant for Gilead Sciences, Merck, Bristol-Myers Squibb, EMD-Serono, and Theratechnologies. K. M. E. has received research grant support from Gilead Sciences and has served as a consultant for Theratechnologies. F. J. P. has served as a speaker and consultant for Gilead Sciences, Janssen, Merck, and Bristol-Myers Squibb. L. P. J. has served as a consultant to Bristol-Myers Squibb. All other authors report no potential conflicts. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.
References
- 1.Gardner EM, Burman WJ, Maravi ME, Davidson AJ. Durability of adherence to antiretroviral therapy on initial and subsequent regimens. AIDS Patient Care STDS 2006; 20:628–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Lima VD, Hogg RS, Harrigan PR et al. Continued improvement in survival among HIV-infected individuals with newer forms of highly active antiretroviral therapy. AIDS 2007; 21:685–92. [DOI] [PubMed] [Google Scholar]
- 3.Gardner EM, Burman WJ, Steiner JF, Anderson PL, Bangsberg DR. Antiretroviral medication adherence and the development of class-specific antiretroviral resistance. AIDS 2009; 23:1035–46. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Viswanathan S, Detels R, Mehta SH, Macatangay BJ, Kirk GD, Jacobson LP. Level of adherence and HIV RNA suppression in the current era of highly active antiretroviral therapy (HAART). AIDS Behav 2015; 19:601–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Palella FJ Jr, Gange SJ, Benning L et al. Inflammatory biomarkers and abacavir use in the Women's Interagency HIV Study and the Multicenter AIDS Cohort Study. AIDS 2010; 24:1657–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Hileman CO, Kinley B, Scharen-Guivel V et al. Differential reduction in monocyte activation and vascular inflammation with integrase inhibitor-based initial antiretroviral therapy among HIV-infected individuals. J Infect Dis 2015; 212:345–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Wada NI, Jacobson LP, Margolick JB et al. The effect of HAART-induced HIV suppression on circulating markers of inflammation and immune activation. AIDS 2015; 29:463–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Lederman MM, Funderburg NT, Sekaly RP, Klatt NR, Hunt PW. Residual immune dysregulation syndrome in treated HIV infection. Adv Immunol 2013; 119:51–83. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Tenorio AR, Zheng Y, Bosch RJ et al. Soluble markers of inflammation and coagulation but not T-cell activation predict non-AIDS-defining morbid events during suppressive antiretroviral treatment. J Infect Dis 2014; 210:1248–59. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.McKibben RA, Margolick JB, Grinspoon S et al. Elevated levels of monocyte activation markers are associated with subclinical atherosclerosis in men with and those without HIV infection. J Infect Dis 2015; 211:1219–28. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Gandhi RT, Sax PE, Grinspoon SK. Metabolic and cardiovascular complications in HIV-infected patients: new challenges for a new age. J Infect Dis 2012; 205(suppl 3):S353–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Borges AH, Dubrow R, Silverberg MJ. Factors contributing to risk for cancer among HIV-infected individuals, and evidence that earlier combination antiretroviral therapy will alter this risk. Curr Opin HIV AIDS 2014; 9:34–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Hsu DC, Sereti I, Ananworanich J. Serious non-AIDS events: immunopathogenesis and interventional strategies. AIDS Res Ther 2013; 10:29. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.El-Sadr WM, Lundgren J, Neaton JD et al. Strategies for Management of Antiretroviral Therapy Study G. CD4+ count–guided interruption of antiretroviral treatment. N Engl J Med 2006; 355:2283–96. [DOI] [PubMed] [Google Scholar]
- 15.Squillace N, Zona S, Stentarelli C et al. Detectable HIV viral load is associated with metabolic syndrome. J Acquir Immune Defic Syndr 2009; 52:459–64. [DOI] [PubMed] [Google Scholar]
- 16.Zhang S, van Sighem A, Kesselring A et al. Episodes of HIV viremia and the risk of non-AIDS diseases in patients on suppressive antiretroviral therapy. J Acquir Immune Defic Syndr 2012; 60:265–72. [DOI] [PubMed] [Google Scholar]
- 17.Podsadecki TJ, Vrijens BC, Tousset EP, Rode RA, Hanna GJ. Decreased adherence to antiretroviral therapy observed prior to transient human immunodeficiency virus type 1 viremia. J Infect Dis 2007; 196:1773–8. [DOI] [PubMed] [Google Scholar]
- 18.Li JZ, Gallien S, Ribaudo H, Heisey A, Bangsberg DR, Kuritzkes DR. Incomplete adherence to antiretroviral therapy is associated with higher levels of residual HIV-1 viremia. AIDS 2014; 28:181–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Pasternak AO, de Bruin M, Jurriaans S et al. Modest nonadherence to antiretroviral therapy promotes residual HIV-1 replication in the absence of virological rebound in plasma. J Infect Dis 2012; 206:1443–52. [DOI] [PubMed] [Google Scholar]
- 20.Konstantopoulos C, Ribaudo H, Ragland K, Bangsberg DR, Li JZ. Antiretroviral regimen and suboptimal medication adherence are associated with low-level human immunodeficiency virus viremia. Open Forum Infect Dis 2015; 2:ofu119. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Kaslow RA, Ostrow DG, Detels R, Phair JP, Polk BF, Rinaldo CR Jr. The Multicenter AIDS Cohort Study: rationale, organization, and selected characteristics of the participants. Am J Epidemiol 1987; 126:310–8. [DOI] [PubMed] [Google Scholar]
- 22.Viswanathan S, Justice AC, Alexander GC et al. Adherence and HIV RNA suppression in the current era of highly active antiretroviral therapy. J Acquir Immune Defic Syndr 2015; 69:493–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Kobin AB, Sheth NU. Levels of adherence required for virologic suppression among newer antiretroviral medications. Ann Pharmacother 2011; 45:372–9. [DOI] [PubMed] [Google Scholar]
- 24.Bangsberg DR, Kroetz DL, Deeks SG. Adherence-resistance relationships to combination HIV antiretroviral therapy. Curr HIV/AIDS Rep 2007; 4:65–72. [DOI] [PubMed] [Google Scholar]
- 25.Cox C, Chu H, Schneider MF, Munoz A. Parametric survival analysis and taxonomy of hazard functions for the generalized gamma distribution. Stat Med 2007; 26:4352–74. [DOI] [PubMed] [Google Scholar]
- 26.Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc Series B Methodol 1995; 57:289–300. [Google Scholar]
- 27.Weiss L, Chevalier MF, Assoumou L et al. Rosuvastatin is effective to decrease CD8 T-cell activation only in HIV-infected patients with high residual T-cell activation under antiretroviral therapy. J Acquir Immune Defic Syndr 2016; 71:390–8. [DOI] [PubMed] [Google Scholar]
- 28.Tien PC, Choi AI, Zolopa AR et al. Inflammation and mortality in HIV-infected adults: analysis of the FRAM study cohort. J Acquir Immune Defic Syndr 2010; 55:316–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Reus S, Portilla J, Sanchez-Paya J et al. Low-level HIV viremia is associated with microbial translocation and inflammation. J Acquir Immune Defic Syndr 2013; 62:129–34. [DOI] [PubMed] [Google Scholar]
- 30.Boyd A, Meynard JL, Morand-Joubert L et al. Association of residual plasma viremia and intima-media thickness in antiretroviral-treated patients with controlled human immunodeficiency virus infection. PLoS One 2014; 9:e113876. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Hatano H, Yukl SA, Ferre AL et al. Prospective antiretroviral treatment of asymptomatic, HIV-1 infected controllers. PLoS Pathog 2013; 9:e1003691. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Rosenblum M, Deeks SG, van der Laan M, Bangsberg DR. The risk of virologic failure decreases with duration of HIV suppression, at greater than 50% adherence to antiretroviral therapy. PLoS One 2009; 4:e7196. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Lima VD, Bangsberg DR, Harrigan PR et al. Risk of viral failure declines with duration of suppression on highly active antiretroviral therapy irrespective of adherence level. J Acquir Immune Defic Syndr 2010; 55:460–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Erlandson KM, Campbell TB. Inflammation in chronic HIV infection: what can we do? J Infect Dis 2015; 212:339–42. [DOI] [PubMed] [Google Scholar]
- 35.Sandler NG, Zhang X, Bosch RJ et al. Sevelamer does not decrease lipopolysaccharide or soluble CD14 levels but decreases soluble tissue factor, low-density lipoprotein (LDL) cholesterol, and oxidized LDL cholesterol levels in individuals with untreated HIV infection. J Infect Dis 2014; 210:1549–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Roff SR, Noon-Song EN, Yamamoto JK. The significance of Interferon-gamma in HIV-1 pathogenesis, therapy, and prophylaxis. Front Immunol 2014; 4:498. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Bermudez EA, Rifai N, Buring J, Manson JE, Ridker PM. Interrelationships among circulating interleukin-6, C-reactive protein, and traditional cardiovascular risk factors in women. Arterioscler Thromb Vasc Biol 2002; 22:1668–73. [DOI] [PubMed] [Google Scholar]
- 38.Pearson CR, Simoni JM, Hoff P, Kurth AE, Martin DP. Assessing antiretroviral adherence via electronic drug monitoring and self-report: an examination of key methodological issues. AIDS Behav 2007; 11:161–73. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Turner BJ. Adherence to antiretroviral therapy by human immunodeficiency virus-infected patients. J Infect Dis 2002; 185(suppl 2):S143–51. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.