icon-    folder.gif   Conference Reports for NATAP  
  17th CROI
Conference on Retroviruses
and Opportunistic Infections
San Francisco CA
February 16-19, 2010
Back grey_arrow_rt.gif
Chronic Inflammation in HIV: CROI 2010
  David H Shepp, MD
Associate Professor of Medicine, Hofstra University School of Medicine Division of Infectious Diseases, North Shore University Hospital-Manhasset

Cardiovascular Disease (CVD).

Pathogenesis research has firmly established a link between chronic inflammation and atherosclerotic CVD in the general population. Untreated HIV infection causes a state of intense chronic inflammation, which accounts it least in part for the higher rates and younger age of onset of cardiovascular disease in HIV infected patients. ART reduces immune activation caused by HIV and many of the markers of chronic inflammation linked to CVD also improve. However, many studies have shown that levels of immune activation and inflammatory markers remain elevated in HIV-infected individuals when compared to the HIV-uninfected, even when ART has produced years of durable viral suppression. This raises the specter of ongoing increased CVD risk even in successfully treated patients. Strategies to further reduce immune activation to normal levels are not established and represent an important unmeet need in HIV treatment research.

In a symposium on Wednesday afternoon, Dr Priscilla Hsue, a cardiologist and associate professor of medicine at the UCSF gave an excellent review of HIV, inflammation and cardiovascular disease. She cited new research presented at this conference exploring the relationship between immune activation/inflammatory markers and markers of CVD risk, the pathogenesis of residual chronic inflammation in patients treated with ART and of potential interventions that might be used in conjunction with traditional ART to further reduce inflammation.

Hsue et al.1 presented data demonstrating rapid progression of carotid intima-media thickness (cIMT) in 285 HIV+ patients as measured by ultrasound at the carotid bifurcation (0.074mm/year) or the internal carotid (0.046mm/year). Both ART treated and untreated patients were studied and in both groups, progression rates were significantly greater than those observed in matched HIV- controls. cIMT reflects deposition of arterial plaque and has been validated as a predictor of cardiovascular disease events in the general population. No difference in progression rates were seen in common carotid, the anatomic location more commonly used to assess cIMT in prior studies. Dr Hsue cited evidence that the carotid bifurcation may be more sensitive to inflammatory changes and thus a better place to measure the vascular effects of HIV+ infection. After adjustment for traditional CVD risk factors, higher high sensitivity c-reactive protein (hsCRP) was independently associated with cIMT thickness at bifurcation, while traditional risk factors correlated better than hsCRP with progression in the common carotid.

Baker et al.2 presented data on cIMT progression in 424 participants in a prospective observational cohort (SUN study). Most were ART treated. Over 2 years, modest (0.013mm) progression in cIMT was observed at the common carotid. Other sites were not assessed, a potential limitation of the study. After adjusting for traditional CVD risk factors, patients with HIV RNA <400 had significantly lower cIMT progression compared to those with >400 (0.011mm v. 0.019 mm; p<0.001). Treatment with an NNRTI also was associated with significantly (p=0.01) lower cIMT progression than treatment with a PI. Significant differences in cIMT progression were not seen when abacavir and tenofovir treatment were compared.

Hsue and colleagues3 also assessed the relation between CVD and inflammation using flow-mediated dilation (FMD), a functional assay for subclinical CVD risk. FMD measures the ability of the brachial artery to dilate in response to endogenous release of nitric oxide following ischemic stress. This cross-sectional study enrolled 98 HIV+ individuals who were virally controlled on ART and 32 HIV-controls. FMD was significantly lower and hsCRP was significantly higher in HIV+ compared to HIV- subjects. hsCRP level was strongly and independently associated with reduced FMD after adjustment for traditional CVD risk factors, a result similar to that seen in the cIMT study. In this study, when exposure to ARVs was examined, abacavir but not protease inhibitors were associated with reduced FMD.

Kaplan et al.4 studied the relation between CVD assessed by cIMT and carotid distensibility and immune activation in HIV+ and HIV- women from the WIHS cohort. 66/116 were on ART. Among HIV+, higher levels of CD4 and CD8 activation(CD38+, HLA-DR+) and senescence markers (CD57+ and/or CD28-) were independently associated with reduced distensibility. Carotid lesions, defined as >1.5mm cIMT, also were associated with T-cell activation and senescence.

Ford et al.5 retrospectively examined the ability of inflammatory markers to predict clinical CVD events among 1892 HIV-infected clinical trials participants at NIAID. Multiple potential CVD inflammatory markers were measured in stored, pre-event plasma from cases who experienced a clinical CVD event and matched controls who did not. The strongest independent association in 4 month pre-event plasma was with elevated d-dimer (p=0.002). In this study, CRP not associated with events. Traditional CVD risk factors such as family history, dyslipidemia, and smoking were also confirmed to increase risk.

These studies provide evidence that CVD, both subclinical changes in arterial morphology or function, and clinical CVD events, correlate with levels of inflammatory markers, and suggest that ART reduces but does not normalize risk attributable to HIV infection. And while the associations described here cannot establish a cause and effect relationship, they do support the hypothesis that chronic inflammation contributes to increased CVD risk in HIV+ individuals, even when responding to ART, and point to the need for further research to understand the mechanism of chronic immune activation and inflammation.

Effects of chronic inflammation on other organ systems and on mortality.

While the link between chronic inflammation and CVD has been more thoroughly investigated, chronic inflammation appears to have an association with disease in other organ systems with overall mortality in HIV-infected persons. These relationships were also explored in a number of studies presented at CROI. Tien et al.6 investigated the relationship between 2 inflammatory markers, CRP and fibrinogen, and mortality in 922 participants followed for 5 years of in the FRAM study. After adjustment for demographic, traditional CVD risk and HIV-related factors, both CRP and fibrinogen levels at baseline were strongly correlated with subsequent mortality. Baseline levels of either marker in the highest tertile were associated with about a 2.6-fold increased risk of death. The effect of elevated CRP or fibrinogen on mortality was significant across all levels of CD4, from <200 to >500.

An analysis of non-AIDS mortality data from the SMART study suggested co-infection with HCV or HBV creates and especially dangerous proinflammatory state that strongly predisposes to death in the absence of continuous ART. A previous analysis showed elevated plasma hyaluronic acid (HA), a correlate of hepatic fibrosis, identified co-infected patients at increased risk of non-AIDS death when randomized to the drug conservation, but not the viral suppression strategy of SMART. In the present analysis, Peters et al.7 found baseline levels of 3 inflammatory markers (hsCRP, IL-6, d-dimer) previously linked to all-cause mortality in the overall SMART study population were higher at baseline in co-infected patients with elevated HA levels, and that IL-6 levels increased dramatically during treatment interruption when baseline HA was elevated. For patients with elevated HA plus an increased level of any of the 3 biomarkers, the adjusted hazard ratio for death increased 4.4-6.1-fold compared to those with normal levels of HA and the biomarker.

To explore the relationship between renal disease and inflammation, Gupta et al.8 reported a small pilot study to define urinary markers of kidney inflammation in patients with HIV and proteinuria but without other known conditions predisposing to proteinuria. Compared to HIV+ but non-proteinuric controls, urinary levels of two inflammatory cytokines, MCP-1 and RANTES (assessed by cytokine/creatinine ratio) were elevated. Studies to determine if these findings reflect systemic or intrarenal inflammation and comparison to HIV + patients with know causes of proteinuric kidney and to HIV - controls are needed to help interpret these findings.

Chronic Inflammation: Pathogenesis & Treatment.

One potential explanation for immune activation and inflammation in patients successfully treated with ART is residual HIV replication below the level of detection of standard plasma HIV RNA assays. Several abstracts reported the results of intensification of standard ART regimens using either raltegravir9-12 or maraviroc13,14. None of these studies found convincing evidence that intensification in patients already well controlled on standard ART produced consistent reduction in HIV RNA or proviral DNA burden, as measured by sensitive assays able to detect HIV at a threshold below that of standard assays, although 2 studies described a transient rise in 2 LTR circles (a sensitive marker of ongoing viral replication)9,13. Additionally, a subset of these studies examined markers of T-cell activation and some, but not all, observed reduced immunologic activation during intensification9,11,13,14, suggesting a direct immunomodulatory effect of ART intensification. These findings were based on small pilot studies and have not yet been shown to affect clinical outcome. None of the studies looking at intensification as a tool to augment CD4 reconstitution found positive effects10,12,14. In aggregate, these studies provide evidence that ongoing HIV replication is unlikely to explain chronic inflammation but leave the door open to a possible benefit from intensification therapy via reduced immune activation.

Previous studies have suggested that translocation of gut microbial flora, as evidenced by elevated plasma levels of bacterial lipopolysaccharide (LPS), may be the driver of chronic immune activation in HIV. The reason for ongoing translocation despite immunologic recovery during ART is not known. Th17 cells play an important role in the immune response to microbial infection and maintenance of mucosal integrity. Favre et al.15 postulated specific defects of Th17 cells could be permissive for microbial translocation. They reported the ratio of Th17 to regulatory T-cells (Treg) was 5-10 fold lower in blood and gut-associated lymphoid tissue (GALT) in HIV+ non-controllers when compared to controllers, patients suppressed on ART and HIV - individuals. The Th17/Treg ratio was highly correlated with both a marker of microbial translocation (16s ribosomal DNA) and T cell activation markers. Evidence was also shown suggesting Th17/Treg abnormalities may be caused by generation of a toxic metabolite of tryptophan metabolism due to increased activity of the enzyme indoleamine dioxygenase in uncontrolled HIV infection.

Another hypothesized cause of chronic immune activation is persistent viral coinfection. A substantial proportion of chronically activated CD8+ T-cells in HIV-infected patients are specific for cytomegalovirus (CMV). In a small (n=30) pilot study, Hunt et al.16 randomized CMV seropositive patients and high levels of activated CD8+ cells despite ART to receive valganciclovir (VGVC) or placebo for 8 weeks. VGCV treatment suppressed CMV infection and reduced CD8+ activation markers significantly when compared to placebo. VGCV was well tolerated over the short course of study treatment. The longer term safety and efficacy of VGCV treatment and the clinical benefit of reduced CD8 activation will need to be established in future trials.

Another potential approach to limiting the effect of HIV infection and concommitant inflammation on CVD is earlier use of ART. Ho et al.17 measured arterial stiffness, another surrogate marker for CVD risk, by carotid/femoral pulse-wave velocity. Patients who initiated ART during acute or early HIV infection in the OPTIONS study were compared to patients in the SCOPE cohort who initiated in the chronic phase when CD4 was less than 350. Arterial stiffness associated with traditional CVD risk factors and after adjustment, with a nadir CD4 <350.


New data presented at CROI 2010 further confirm the importance of chronic inflammation in the pathogenesis of non-AIDS complications of HIV infection, even in those successfully treated with ART. The role in of chronic inflammation in CVD pathogenesis is better studied and better established than in complications affecting other organ systems. Despite some intriguing hints into the pathogenesis and treatment of chronic inflammation provided at this conference, the cause remains unclear and is likely multifactorial. Studies of treatment strategies designed to directly address chronic inflammation are beginning to appear. Further research in this area is needed to improve life-expectancy and reduce comorbidity in HIV-infected persons treated with long-term ART.

1. Hsue P, et al. Rapid progression of atherosclerosis at the carotid bifurcation is linked to inflammation in HIV-infected patients. CROI San Francisco, Feb 25-29, 2010, abstract 125.

2. Baker J, et al. Progression of carotid intima-media thickness in a contemporary HIV cohort. CROI, San Francisco, Feb 25-29, 2010, abstract 126.

3. Hsue P, et al. Inflammation is associated with endothelial dysfunction among individuals with treated and suppressed HIV infection. CROI, San Francisco, Feb 25-29, 2010, abstract 708.

4. Kaplan J, et al. T cell senescence and T cell activation predict carotid atherosclerosis in HIV-infected women. CROI, San Francisco, Feb 25-29, 2010, abstract 709.

5. Ford E, et al. Elevated d-dimer but not CRP levels in HIV+ patients prior to incident myocardial infarction or other cardiovascular disease event. CROI, San Francisco, Feb 25-29, 2010, abstract 713.

6. Tien P, et al. Inflammation and mortality in HIV-infected adults: Analysis of the FRAM study cohort. CROI, San Francisco, Feb 25-29, 2010, abstract 725.

7. Peters L, et al. Biomarkers of inflammation and coagulation and risk of non-AIDS death in HIV/hepatitis co-infected patients in the SMART study. CROI, San Francisco, Feb 25-29, 2010, abstract 660.

8. Gupta S et al. Intrarenal inflammation contributes to HIV-related proteinuria. CROI, San Francisco, Feb 25-29, 2010, abstract 736.

9. Buzon MJ, et al. HIV-1 replication and immune dynamics are impacted by raltegravir intensification of HAART-suppressed patients. CROI, San Francisco, Feb 25-29, 2010, abstract 110LB.

10. Hatano H, et al. Raltegravir intensification in antiretroviral-treated patients exhibiting a suboptimal CD4+ T cell response. CROI, San Francisco, Feb 25-29, 2010, abstract 101 LB.

11. Yukl S, et al. Effect of raltegravir-containing intensification on HIV burden and T cell activation in the gut of HIV+ adults on suppressive ART. CROI, San Francisco, Feb 25-29, 2010, abstract 279.

12. Wiegand A, et al. Raltegravir intensification does not reduce persistent HIV-1 viremia in treatment-experienced patients. CROI, San Francisco, Feb 25-29, 2010, abstract 280

13. Gutierrez C, et al. Effect of intensification with a CCR5 antagonist on the decay of the HIV-1 latent reservoir and residual viremia. CROI, San Francisco, Feb 25-29, 2010, abstract 284

14. Wilkin T, et al. Maraviroc intensification for suboptimal CD4+ cell response despite sustained virologic suppression: ACTG 5256. CROI, San Francisco, Feb 25-29, 2010, abstract 285

15. Favre D, et al. Trytophan catabolism by 2,3 indoleamine dioxygenase alters the balance between Th17 and T regulatory cells in progressive HIV infection. CROI, San Francisco, Feb 25-29, 2010, abstract 94LB.

16. Hunt P, et al. Valganciclovir reduces CD8+ T cell activation among HIV-infected patients with suboptimal CD4+T cell recovery during ART. CROI, San Francisco, Feb 25-29, 2010, abstract 380.

17. Ho et al. Earlier initiation of ART in HIV-infected individuals Is associated with reduced arterial stiffness. CROI, San Francisco, Feb 25-29, 2010, abstract 707