October 28, 2001
Metabolic problems remain a major concern surrounding the use of highly active antiretroviral therapy (HAART). One of these, hyperlipidemia, occurs commonly in patients on protease inhibitor (PI)-based HAART, many of whom may experience elevations in their cholesterol, LDL cholesterol and triglycerides. While the clinical significance of these elevations remains unclear, with some studies indicating that protease inhibitors are associated with increased coronary artery lesions and cardiovascular mortality and others not, this problem clearly is on the minds of many clinicians and HIV-infected patients.
In a satellite conference sponsored by Boehringer-Ingelheim at the 8th European Conference on Clinical Aspects and Treatment of HIV Infection (ECCATHI), lipid problems associated with HAART were reviewed and the more favorable lipid profiles, associated with the usage of non-nucleoside reverse transcriptase inhibitors (NNRTIs), in particular nevirapine, were discussed.
Peter Reiss reviewed data from the lipid sub-study of the Atlantic Trial that indicates that nevirapine-based HAART is associated with only modest increases in total cholesterol, LDL cholesterol (the "bad" cholesterol) and triglycerides, but a significant increase in HDL cholesterol (the "good" cholesterol), which results in an improvement in the total cholesterol:HDL ratio. This effect has been shown to be sustained through 96 weeks of treatment. The 96-week data from this study were presented in a poster presentation at this conference.1
While there were no significant differences in patient baseline lipid profiles between the three treatment arms of the Atlantic Trial (didanosine plus stavudine, in combination with either indinavir, nevirapine or lamivudine) at 24 weeks, all arms had a rise in total cholesterol and LDL cholesterol with the lowest increases in the lamivudine arm. But in the nevirapine arm, HDL rose 49% and HDL size and large HDL number also increased, all of which was greater than in the other two arms. Due to this rise in HDL, the total cholesterol:HDL ratio, perhaps the best indicator of cardiovascular risk, decreased 14% in the nevirapine arm, 9% in the lamivudine arm, and was close to unchanged in the indinavir arm of the trial. Further, apolipoprotein-A1, the protein produced by the liver and which is "loaded" with cholesterol to form HDL cholesterol, also increased more in the nevirapine arm than the other arms of the trial, which may be an early indication of the etiology of the HDL increases associated with nevirapine. A multivariate analysis, which controlled for other potential factors that may have affected lipids, including CD4+ T-cell counts and viral load, found only the nevirapine treatment to be significant.
In a small subset of the patients examined, the lipid effects associated with nevirapine-based HAART appear to be sustained to 48 and 96 weeks, with HDL increases in the 40% range in the nevirapine arm at 96 weeks of therapy versus more modest changes of 20% and 6% in the lamivudine and indinavir arms, respectively. As a result, at the end of 96 weeks of therapy, the total cholesterol:HDL ratio decreased by 6% in the nevirapine arm, while it increased 2% and 25% in the lamivudine and indinavir arms, respectively. The total cholesterol, LDL cholesterol, and triglyceride changes were similar between the three arms at 96 weeks.
Bonaventura Clotet presented data that corroborates the effect upon lipids whether the patients who start on nevirapine are naive or protease inhibitor-experienced. In the Barcelona protease inhibitor switch study -- a relatively small study involving only 77 patients -- patients who switched from a protease inhibitor to nevirapine or efavirenz maintained virologic suppression and immunologic benefit, while in the nevirapine arm, patients experienced an improvement in their lipid profile similar to that seen in the Atlantic Trial data. These improvements were also seen in the efavirenz arm in which lipid parameters improved somewhat as well but not as impressively as in the nevirapine arm. There was no change in either of the arms as far as body fat changes.
The presenters at this conference conceded that these findings are likely to be in large part an NNRTI class effect, although the total cholesterol/HDL cholesterol change is not quite as impressive in patients on efavirenz-based HAART. Furthermore, the presenters admitted that the clinical significance of these findings remains unclear and that they have no data to support relevance in the "setting of HIV treatment." Whether these lipid changes are significant and ultimately affect clinical outcomes will need to be determined by other studies that take into account the effect of these and other risk factors for coronary artery disease, the use of medications to treat and control lipid problems associated with protease inhibitors, the nature and duration of the lipid changes, and numerous other factors which may impact cardiovascular disease and mortality in HIV-infected patients on and off antiretroviral therapy. Still, in certain patient classes, lipid improvements may be important and these data may be considered useful in guiding treatment decisions.