Abacavir May Increase Risk for CVD in Subset of Patients

Bob Roehr

August 12, 2008

August 12, 2008 (Mexico City) — Debate is ongoing on whether use of the nucleoside reverse transcriptase inhibitors (NRTIs) abacavir (ABC) and didanosine (ddI) heightens the risk for cardiovascular disease (CVD). It came to the fore in April in a paper published in The Lancet that analyzed data from the observational D:A:D cohort study, and it continued here at the XVII International AIDS Conference.

Jens Lundgren, MD, from the University of Copenhagen, Denmark, and colleagues wanted to know whether the association of excess risk for myocardial infarction (MI) that they found in analysis of D:A:D "could be reproduced in another data set where utilization of various NRTIs differed from that of D:A:D."

They turned to the SMART study, a randomized trial comparing continuous and guided interruption of antiretroviral therapy. The researchers stratified 4544 patients into 3 categories according to use of NRTIs: ABC but not ddI, ddI with and without ABC, and use of other NRTIs. Baseline characteristics of the groups were similar, including risk factors for cardiovascular disease, diabetes, current smoking, and use of hypertension and lipid-lowering medications.

The analysis evaluated 4 strata of CVD: CVD major; MI; CVD, expanded; and CVD minor. The study found an almost exact overlap of the adjusted and unadjusted hazard ratio in every category that favored "other" more than ABC: 1.8, 4.3, 1.9, and 2.7 in each of the 4 categories of CVD, respectively. Similar patterns of association were seen when use of ABC was compared with use of tenofovir.

Evaluation of a subset of patients with biomarker data available at baseline found high-sensitivity C-reactive protein (hs-CRP) and interleukin 6 (IL-6) levels to be 27% (P = .02) and 16% (P = .02) higher for patients receiving ABC (n = 175) compared with other NRTIs (n = 500). Use of ddI was not associated with altered risk for CVD, nor with altered levels of biomarkers.

Dr. Lundgren concluded, "Consistent with D:A:D, current use of abacavir during follow-up in SMART was associated with an excess risk of CVD. And furthermore, abacavir use at study entry was associated with increased levels of IL-6 and hs-CRP.

"The drug dose does not appear to affect the underlying atherosclerotic process per se but may cause coronary arthritis that leads to instability of plaques. This appears to be only clinically relevant to consider among patients with elevated underlying cardiovascular risk."

Responding to a question, Dr. Lundgren said this excess risk is seen in patients who have been receiving ABC for an extended period of time, so it is unlikely that it is linked to the HLA 5701 phenotype. He encouraged researchers to look for other haplotypes that might contribute to "ongoing, maybe subclinical inflammatory elevation."

Dr. Lundgren said that there appeared to be no relationship between the use of protease inhibitors and increased cardiovascular risk in either the D:A:D or SMART studies.

A clinician from Florida said that he has many patients on ABC who by all accounts seem to be doing well. He was concerned as to whether he should switch them to a different regimen.

Dr. Lundgren noted that the clinical relevance of these studies is likely to apply only to a subset of patients, "namely, those who have several underlying cardiovascular risk factors. For most patients, this is not an issue."

GlaxoSmithKline's Response

GlaxoSmithKline, reacting to the earlier publication of the D:A:D analysis, drew upon its HIV Data Repository to review and conduct a composite analysis of data from the 14,683 participants who had been in 54 clinical trials for 24 weeks or more, and who had been exposed to abacavir. Earlier in the week, Jaime Hernandez, MD, from GlaxoSmithKline, presented that analysis.

The data showed no difference between those who had and those who had not been exposed to ABC. Baseline patient characteristics were similar, as were treatment experience, lipids, and glucose values. Indices of CVD were low and comparable, with the ABC-exposed patients having a 0.249% frequency of coronary artery disorders and those who were not exposed having a 0.416% frequency. The rates of coronary artery disease per 1000 patient-years were 3.447 and 5.817, respectively (P = .055). The rates of MI were 2.039 vs 2.2363 (P = .706).

Dr. Hernandez acknowledged the limitations of the initial drug development studies in possibly capturing long-term and low-incidence events such as enhanced risk for cardiovascular disease. Given the transition of HIV disease management from that of an acute to a chronic condition, he said, "In future studies we need to incorporate cardiovascular risk factors and other biomarker collections of all non-AIDS risk factors."

Potential Reasons for Different Trial Results

During discussion, Cornell University researcher Roy Gulick, MD, noted that "there are inherent differences between clinical trials data and human cohort data," particularly with regard to exclusionary criteria for the trials. This might help to explain the discordance between the clinical trials and cohort analysis.

Another researcher commented that participants in the drug development trials were, on average, younger than those in the cohort studies, and that risk for CVD increases with age. She asked whether the company had done a subset analysis of older trial participants. Dr. Hernandez said it has not.

Kenneth Mayer, MD, a researcher at Brown University, Providence, Rhode Island, and Fenway Community Health, Boston, Massachusetts, told Medscape HIV/AIDS that this information reinforces the need to monitor patients for cardiovascular risk factors: "If you work on those, you can minimize any potential risk from abacavir, if it really does exist." He called abacavir "a good drug" and said that he would hate to see an overreaction on the part of clinicians who would be afraid to use it where appropriate.

The D:A:D and SMART studies were funded by government health agencies in the United States and Europe. The GlaxoSmithKline analysis was conducted by that company. Dr. Hernandez is an employee of GlaxoSmithKline. Dr. Lundgren, Dr. Gulick, and Dr. Mayer have disclosed no relevant financial relationships.

XVII International AIDS Conference: Abstract THAB0305, presented August 7, 2008; abstract WEAB0106, presented August 6, 2008.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.