Data from Multiple Studies -- D:A:D, ANRS CO4, ALLRT, STEAL, HEAT -- Fail to Resolve Abacavir (Ziagen) Cardiovascular Risk Issue

By Liz Highleyman

The latest data from several studies presented at the 16th Conference on Retroviruses and Opportunistic Infections (CROI 2009) this month in Montreal -- some focusing on clinical outcomes and some looking at biomarkers -- have not yet resolved the ongoing debate about whether abacavir (Ziagen, also in the Epzicom and Trizivir coformulation pills) increases the risk for cardiovascular disease in HIV patients.

Investigators with the D:A:D (Data Collection on Adverse Events of Anti-HIV Drugs) study team sparked the debate at last year's Retrovirus conference, when they reported that recent use of abacavir increased the risk of myocardial infarction (MI) by 90%; in addition, didanosine (ddI, Videx) increased the risk by 49%.

At last summer's International AIDS Conferenc, researchers with the large SMART treatment interruption study also reported that abacavir was associated with a heightened risk of cardiovascular disease, but a pooled analysis of 54 clinical trials by abacavir manufacturer GlaxoSmithKline (GSK) did not find an increase in heart attacks or other types of cardiovascular events.

Since then, numerous research teams have gone back to reanalyze old data or have started new studies to shed more light on the question.

D:A:D Study

At this year's meeting, Jens Lundgren presented longer follow-up data from D:A:D, a multinational collaboration involving 11 prospective cohorts from the U.S., Europe, and Australia. The latest analysis included data from 33,308 HIV positive participants followed through February 2008, for a total of 178,835 person-years of follow-up.

The researchers used Poisson regression to assess the impact of specific antiretroviral drugs from 3 classes -- 7 nucleoside/nucleotide reverse transcriptase inhibitors (NRTIs), 2 non-nucleoside reverse transcriptase inhibitors (NNRTIs), and 4 protease inhibitors (PIs) -- on heart attack risk. D:A:D only reports findings for drugs used by a large number of patients over a considerable length of time. Tenofovir (Viread, also in the Truvada and Atripla coformulations) was omitted last year, but included in the present analysis.

Data were adjusted for demographic characteristics, HIV disease status, blood lipid levels, metabolic measurements, and known cardiovascular risk factors such as smoking and family history. About 20% of follow-up time was contributed by patients with moderate to high Framingham cardiovascular risk scores.

Results

During the follow-up period, 580 participants experienced an MI.

There was no observed association between MI risk and recent or cumulative exposure to the NNRTIs efavirenz (Sustiva) or nevirapine (Viramune).

There was also no link between MI risk and 2 studied PIs, nelfinavir (Viracept) and saquinavir (Invirase).

There was a significant association between increased MI risk and cumulative -- but not recent -- exposure to the PIs indinavir (Crixivan) and lopinavir/ritonavir (Kaletra) (RR 1.12 and 1.13 per year, or increases of 12% and 13%, respectively).

The PI association was smaller after adjusting for blood lipid levels, but did not change after adjusting for ritonavir boosting.

5 of the studied NRTIs demonstrated no significant associations with MIs.

However, the previously reported association between MIs and recent use of abacavir (RR 1.68, or 68% increase) and ddI (RR 1.41, or 41% increase) remained.

In this longer analysis, the excess risk associated with recent abacavir exposure appeared to increase with longer use.

CD4 cell count and HIV viral load did not predict MI risk.

Noting the difference in risk profiles within drug classes, Lundgren said it was difficult to determine a "class effect," and emphasized the need to examine specific drugs rather than broad classes.

With regard to "channeling bias" (the tendency to prescribe abacavir to patients with pre-existing cardiovascular disease, since it was thought to be safer than thymidine analog NRTIs), he noted that if such a bias was present, it should have also disfavored tenofovir, which was not the case.

Finally, given the lack of association between heart attack risk and CD4 count or viral load, he suggested that factors other than immune suppression and ongoing HIV replication probably underlie the link, perhaps including an inflammatory reaction that affects blood vessel walls and may promote pre-existing atherosclerosis.

ANRS CO4

In an accompanying presentation, Dominique Costagliola described a case-control study nested with ANRS CO4, looking at the effect of specific antiretroviral drugs on MI risk among more than 11,500 patients in the French Hospital Database -- an analysis undertaken after last year's D:A:D report.

The 289 HIV patients in the database who experienced a confirmed first MI between January 2000 and December 2006 were matched with up to 5 control subjects (total 884) of the same sex and similar age who had never had a heart attack. Again, the researchers controlled for cardiovascular risk factors and HIV disease status.

The investigators examined associations between heart attack risk and recent (past 1 year) or cumulative use of 8 NRTIs and 5 PIs (other drugs were analyzed, but statistical analysis was only reported for those used by at least 100 patients.)

Results

No association with MIs was observed for saquinavir.

Cumulative exposure to the other studied PIs was linked to increased MI risk, but this was only statistically significant for lopinavir/ritonavir (OR 1.37 per year) and amprenavir (Agenerase) or fosamprenavir (Lexiva) considered together as 1 drug (OR 1.52 per year).

Among the NRTIs, no association was seen with lamivudine (3TC, Epivir), emtricitabine (Emtriva), tenofovir or -- in contrast to D:A:D -- ddI.

There was a trend toward increased risk with the thymidine analogs zidovudine (AZT, Retrovir) and stavudine (d4T, Zerit), but this did not reach statistical significance.

Patients with recent abacavir exposure had about twice the risk of having a heart attack than unexposed individuals (OR 1.97).

Participants who had used abacavir in the past but stopped before 1 year (for example, due to adverse events) had a slight but non-significant increase in risk (OR 0.31), but this was not the case for patients who remained on the drug.

There was no observed interaction between abacavir exposure and other cardiovascular risk factors (again, in contrast to D:A:D).

"In our study, we found that the risk of MI was increased by cumulative exposure to [lopinavir/ritonavir] and to amprenavir or fosamprenavir," the researchers concluded. "Initiating abacavir was also associated with an increased risk of MI while longer exposure to abacavir was not."

ALLRT (ACTG A5001)

Constance Benson and colleagues with the AIDS Clinical Trials Group ALLRT study (ACTG Longitudinal Linked Randomized Trials, aka ACTG A5001) likewise conducted an analysis of cardiovascular events following the 2008 D:A:D report.

The present analysis included more than 3205 HIV patients who initiated first-line treatment in 5 ACTG randomized clinical trials of antiretroviral therapy (ART); 781 started a regimen containing abacavir. The researchers recorded 63 total severe cardiovascular events, including 27 MIs.

"In contrast to D:A:D and SMART, we did not find a significant association between recent abacavir use and MI or severe cardiovascular disease risk for ART-naive patients randomized to an initial abacavir regimen," the investigators concluded. "Our results suggest the association with recent abacavir use in other studies may be a marker for other factors not discerned in their analyses."

STEAL

In the Australian STEAL study, David Cooper, Andrew Carr, and colleagues studied 360 participants who simplified their antiretroviral regimen by switching from individual NRTIs to either the Epzicom or Truvada coformulation (20% and 30%, respectively, were taking abacavir and tenofovir as individual NRTIs before switching to the combo pills).

This study found that patients who switched to Epzicom and Truvada were equally likely to maintain good viral suppression at 96 weeks, but those taking a abacavir-containing regimen had a higher rate of cardiovascular disease events -- 7 total, including 3 MIs, 1 stroke, and 1 coronary artery bypass -- than those taking a tenofovir-containing regimen (1 cardiovascular event), though the difference did not reach statistical significance due to the small numbers involved.

WIHS and MACS

Several other researchers looked at various biomarkers associated with cardiovascular risk. Changes in biomarkers for inflammation, coagulation (clotting), and other related processes may help predict cardiovascular events, and can also help clarify the mechanisms underlying associations between these events and specific drugs.

Frank Palella presented late-breaking findings from a combined analysis of inflammatory markers in HIV positive participants in the Women's Interagency HIV Study (WIHS) and the Multicenter AIDS Cohort Study (MACS).

The analysis included 197 women in WIHS and 129 gay/bisexual men in MACS who either took abacavir as part of their first-line regimen or switched to it later, matched 1:1 with patients who didn't take the drug. The investigators measured blood levels of high-sensitivity C-reactive protein (hsCRP), interleukin 6 (IL-6), and D-dimer at baseline and at the first visit after starting abacavir (index visit) or a matched visit if abacavir unexposed.

After adjusting for potential confounding factors including demographics and cardiovascular risk factors, abacavir users had a 7% lower D-dimer level, 11% higher hsCRP level, and 5% lower IL-6 level than non-users, none of which were statistically significant. Compared with men, women had significantly lower hsCRP (33% less) and IL-6 (38% less).

Looking at changes between baseline and the index visit (mean of about 4 years), patients experienced overall decreases in D-dimer and IL-6, but the changes between visits were not significantly different for the abacavir and non-abacavir groups. hsCRP also fell in both arms, but more so in the abacavir unexposed patients.

Based on these findings, the researchers concluded, "Abacavir use was not independently associated with elevated plasma levels of hsCRP, IL-6, and D-dimer."

Platelet function

In another late-breaker, Claudette Satchell reported on a study of platelet function in patients exposed or unexposed to abacavir. Platelets, or thrombocytes, are blood cell fragments necessary for clotting.

The study included 30 patients taking abacavir and 28 taking other NRTIs seen at Mater Misericordiae Hospital in Dublin, Ireland. Patients were matched for demographics and HIV disease status; 4 in the abacavir group and 8 in the unexposed group had a history of cardiovascular disease.

Although platelets showed hyper-reactivity (indicating increased propensity for clotting) in HIV positive individuals in general (as reported in a poster at the conference), this was more pronounced among patients taking abacavir. Increased platelet reactivity was significantly greater among abacavir recipients when exposed to 3 clot-stimulating agents (adenosine diphosphate, epinephrine, and collagen).

The researchers concluded that this platelet hyper-reactivity may help explain increased rates of MI in abacavir-treated patients.

HEAT

Finally, in a poster presentation, Grace McComsey and colleagues described an analysis of biomarkers of inflammation and endothelial activation among 476 treatment-naive participants in the HEAT study.

HEAT, sponsored by GSK, was a prospective, randomized head-to-head trial comparing Epzicom against Truvada. The researchers measured blood levels of hsCRP, IL-6, and vascular cell adhesion molecule-1 (sVCAM-1) at baseline and again at weeks 48 and 96.

Levels of hsCRP, IL-6, and sVCAM-1 decreased after starting therapy in both groups, and reductions were not significantly different between the 2 arms for any of the biomarkers. Because there were so few cardiovascular events overall, the investigators were not able to compare rates in the 2 arms.

"These data do not suggest that [Epzicom] or [Truvada] contribute to an increase in cardiovascular risk mediated by inflammation or worsening endothelial activation," the researchers concluded. "The findings from this randomized, prospective data do not support the hypothesis of increased inflammation attributed to abacavir from recent observational, cohort studies."

Overview

Following the Palella and Satchell presentations, Peter Reiss from the University of Amsterdam gave an overview and summary of research to date about the cardiovascular risk of abacavir.

Weighing in favor of an increased risk associated with recent abacavir use are D:A:D, SMART, ANRS CO4, and possibly STEAL.

On the other side, finding no association, are the GSK 54-study analysis, ALLRT, WIHS/MACS, HEAT, and the Spanish BICOMBO study.

In attempting to make sense of these inconsistent findings, Reiss indicated that reviewing the specific characteristics of the various patient populations under study might offer some clues to the disparities. Participants in studies that have not seen an increased risk, for example, tend to be 7-10 years younger, and it might be harder to see an effect in a population with a lower number of cardiovascular events.

In addition, he suggested, "one could for instance speculate whether the likelihood of identifying the cardiovascular disease risk associated with abacavir may be greater in those who are first exposed after their HIV infection is already suppressed" -- that is, people who switch to abacavir versus treatment-naive patients.

Furthermore, he added, "the data suggest a pathogenic mechanism (possibly of a proinflammatory nature) that involves acute processes, such as plaque rupture or subsequent thrombosis," rather than a chronic process such as initial plaque (atheroma) formation.

One puzzling question, he noted, is why abacavir appears to be associated with heart attacks, but not strokes, given that these are caused by the same process of arteries becoming blocked by plaque and ensuing clot formation.

Another important point is that while large cohort studies such as D:A:D and SMART found large increases in the relative risk of cardiovascular events associated with abacavir use, absolute numbers of such events have been small, especially among people with no other cardiovascular risk factors.

For now, Reiss concluded, "it seems prudent to withhold abacavir from patients with high underlying cardiovascular disease risk if suitable alternative regimens are available. If not, patients' absolute cardiovascular disease risk in the presence of abacavir should be minimized by aggressive management of traditional cardiovascular risk factors" -- with a particular emphasis on smoking cessation.

2/20/09

References

J Lundgren, P Reiss, S Worm, and others. Risk of Myocardial Infarction with Exposure to Specific ARV from the PI, NNRTI, and NRTI Drug Classes: The D:A:D Study. 16th Conference on Retroviruses and Opportunistic Infections (CROI 2009). Montreal, Canada. February 8-11, 2009. Abstract 44LB.

S Lang, M Mary-Krause, L Cotte, and others. Impact of Specific NRTI and PI Exposure on the Risk of Myocardial Infarction: A Case-Control Study Nested within FHDH ANRS CO4. CROI 2009. Abstract 43LB.

C Benson, H Ribaudo, E Zheng, and others. No Association of Abacavir Use with Risk of Myocardial Infarction or Severe Cardiovascular Disease Events: Results from ACTG A5001. CROI 2009. Abstract 721.

D Cooper, M Bloch, A Humphries, and others. Simplification with Fixed-dose Tenofovir/Emtricitabine or Abacavir/Lamivudine in Adults with Suppressed HIV Replication: The STEAL Study, a Randomized, Open-label, 96-Week, Non-inferiority Trial. CROI 2009. Abstract 576.

F Palella, S Gange, R Elion, and others. Inflammatory Markers among Abacavir and non-Abacavir Recipients in the Womens Interagency HIV Study and the Multicenter AIDS Cohort Study. CROI 2009. Abstract 150LB.

C Satchell, E O'Connor, A Peace, and others. Platelet Hyper-Reactivity in HIV-1-infected Patients on Abacavir-containing ART. CROI 2009. Abstract 151LB.

G McComsey, K Smith, P Patel, and others. Similar Reductions in Markers of Inflammation and Endothelial Activation after Initiation of Abacavir/Lamivudine or Tenofovir/Emtricitabine: The HEAT Study. CROI 2009. Abstract 732.

P Reiss. Abacavir and Cardiovascular Risk. CROI 2009. Abstract 152.