CROI: Biomarker changes may help explain detrimental effects of treatment interruption

This article is more than 16 years old. Click here for more recent articles on this topic

Several biomarkers associated with inflammation, blood coagulation, or endothelial dysfunction are significantly elevated during treatment interruption and may partially explain the increased risk of cardiovascular disease and death seen in those who interrupted antiretroviral therapy during the SMART study, according to a presentation at the Fifteenth Conference on Retroviruses and Opportunistic Infections today in Boston.

D-dimer and IL-6 linked to increased heart disease and mortality in SMART

SMART was a large international treatment strategy trial comparing CD4 cell-guided structured treatment interruption versus continuous antiretroviral therapy, further details of which have been extensively reported on aidsmap, including some from this conference.

SMART’s headline finding, first reported at the 2006 Retroviruses conference, was that individuals in the treatment interruption arm had a higher rate of opportunistic disease or death, as well as an unexpected increase in presumably non-AIDS-related serious heart, kidney, and liver problems. According to the researchers, there was a “marginally statistically significant raised risk of major cardiovascular events” in those who interrupted therapy. In January 2006, the study was halted and all participants were urged to stay on continuous therapy.

To further elucidate the possible mechanisms underlying this increased risk, Lewis Kuller from Department of Epidemiology at the University of Pittsburgh presented data on behalf of the SMART Study Group from an analysis of six biomarkers that have been associated with mortality and cardiovascular disease in the general population.



Genes, proteins or chemicals that can act as signals for certain diseases.

treatment interruption

Taking a planned break from HIV treatment, sometimes known as a ‘drugs holiday’. As this has been shown to lead to worse outcomes, treatment interruptions are not recommended. 


D-dimer is one of the protein fragments produced when a blood clot gets dissolved in the body. In case of inflammation (for example, due to HIV) its level in the blood can significantly rise, as inflammation initiates clotting and decreases the activity of natural anticoagulant mechanisms.


Relating to the heart and blood vessels.


A group of diseases characterized by high levels of blood sugar (glucose). Type 1 diabetes occurs when the body fails to produce insulin, which is a hormone that regulates blood sugar. Type 2 diabetes occurs when the body either does not produce enough insulin or does not use insulin normally (insulin resistance). Common symptoms of diabetes include frequent urination, unusual thirst and extreme hunger. Some antiretroviral drugs may increase the risk of type 2 diabetes.

In a nested case-control analysis, the researchers paired each of the 85 individuals who died in SMART through January 2006 (55 in the treatment interruption arm and 30 in the continuous therapy arm) with two surviving participants of the same sex, age, and country.

The participants who died had more health risks than survivors, even before randomisation in the study. At baseline, the patients who died had a lower CD4 count (545 vs. 614 cells/mm3), were more likely to smoke, and had higher more diabetes, high blood pressure, and pre-existing heart disease. HIV viral load and use of HAART, however, did not differ.

The investigators assessed a wide range of demographic, cardiovascular, and HIV-related variables, looking at how these might be related to risk of death.

The current analysis focused on six biomarkers. Four of these reflect inflammation: high sensitivity-C reactive protein (hsCRP), interleukin-6 (IL-6), serum amyloid A, and amyloid P. The two others are measures of blood coagulation, or clotting: D-dimer and prothrombin fragment 1+2.

Comparing baseline levels of these markers in SMART participants who died and those who survived, the researchers found that two biomarkers were associated with a large increase in the likelihood of death. After factoring in other variables, the adjusted odds ratios (OR) was 11.8 for IL-6 and 26.5 for D-dimer. They also found that the risk of death increased in a linear fashion as D-dimer levels increased.

Hs-CRP and amyloid P were associated with smaller increases in mortality risk.

In another type of analysis (not presented in detail at the conference), the investigators also looked at stored blood samples from a random subset of about 250 individuals in each of the study arms, taken before and one month after randomisation.

They found that the level of IL-6 increased by about 30% and D-dimer increased by about 15% in the treatment interruption arm, while remaining stable in the continuous therapy arm. Both markers increased progressively as HIV viral load rose after stopping treatment.

Not only baseline levels of D-dimer and IL-6, but also increases in these markers after interrupting antiretroviral therapy, were associated with a rising risk of death.

Dr Kuller said that that the link between increased levels of D-dimer and IL-6 and mortality risk was “extraordinary,” far above the typical associations usually seen in general population studies of cardiovascular risk factors. The odds ratios, especially for D-dimer, were “truly remarkable,” he added.

Interestingly, increased levels of IL-6 and D-dimer were more strongly associated with all-cause mortality than with heart-related death specifically, indicating that their detrimental effects likely extend beyond the cardiovascular system.

“These data, the researchers concluded, “suggest that HIV infection results in activation of coagulation and inflammatory pathways that may impact multiple organs.” They suggested that HIV viraemia may directly affect the vascular endothelium, or lining of the blood vessels, resulting in increased coagulation.

Calling these findings “very clinically relevant,” Dr Kuller added that the study raises the question of whether these biomarkers should be monitored in people with HIV on antiretroviral therapy.

STACCATO finds raised biomarkers for endothelial dysfunction during treatment interruption

Data from a much smaller CD4 cell-guided treatment interruption study, STACCATO, results of which have previously been reported here, added strength to the SMART team’s hypothesis.

Dr Alexandra Calmy of Geneva University Hospital in Switzerland presented data from a study that assessed measurements of endothelial dysfunction among participants in the STACCATO study.

Endothelial dysfunction refers to reduced flexibility in the lining of the blood vessels, and is an early indicator of hardening and narrowing of the arteries (atherosclerosis), one of the first detectable signs of cardiovascular disease.

In STACCATO, all participants were started on antiretroviral therapy for six months before randomisation to the structured treatment interruption or continuous therapy arms. The current analysis included 145 Thai participants (97 in the interruption arm and 48 in the continuous treatment arm).

Compared with the SMART study population, more STACCATO participants were women (62%), they had more advanced HIV disease with a lower average CD4 cell count, and had lower cardiovascular risk, with lower baseline lipid levels, less diabetes, and few smokers.

The investigators analysed blood samples collected before initiating antiretroviral therapy, at the time of randomisation, at week 12 (while patients who interrupted were off therapy), and (for the interrupters) three months after resuming therapy.

A variety of biomarkers were measured including markers of endothelial dysfunction (s-VCAM-1 and p-selectin), inflammatory markers (CRP, IL-6, IL-10, MCP-1, and MIP 1-alpha), and adipokines (leptin and adiponectin).

At 12 weeks after randomisation, changes in biomarkers differed in the continuous therapy arm and the treatment interruption arm. These differences were significant for three of the markers, MCP-1, MIP 1-alpha, and adiponectin.

Levels of MCP-1 and s-VCAM-1 increased after antiretroviral therapy was interrupted, followed by a slight but non-significant decrease after treatment resumed. IL-10 and adiponection fell during treatment interruption, but did not rise after patients went back on treatment. These biomarker changes were strongly associated with changes in HIV viral load. Interestingly, this study—unlike the SMART analysis—did not find a link between viral load and IL-6.

None of the participants in this STACCATO study population experienced cardiovascular events or heart-related death, so the researchers were unable to look for associations between the biomarkers and cardiovascular clinical outcomes.

Dr Calmy concluded that the research team observed significant differences in biomarkers of endothelial activation between individuals on treatment and those who had interrupted treatment, and that there was only incomplete reversal after restarting therapy.

These results may help explain findings from the SMART study reported on Monday showing that intermittent antiretroviral therapy interruption is associated with long-term consequences that persist after resuming treatment.


Kuller L et al. Elevated levels of interleukin-6 and D-dimer are associated with an increased risk of death in patients with HIV. Fifteenth Conference on Retroviruses and Opportunistic Infections, Boston, abstract 139, 2008.

Calmy A et al. HIV activates markers of cardiovascular risk in a randomized treatment interruption trial: STACCATO. Fifteenth Conference on Retroviruses and Opportunistic Infections, Boston, abstract 140, 2008.