• Prospective Study of Hyperlipidemia in ART-Naive Subjects Taking Abacavir/Combivir (ABC/COM), COM/nelfinavir (NFV), or Lamivudine (3TC)/Stavudine (d4T)/NFV (GSK Protocol ESS40002)
  Presenter: Princy Kumar

Further Evidence of Metabolic Risks Lower with PI-Sparing Regimens

Most of the research on metabolic disturbances, at least during the first few months of therapy, points to a role by protease inhibitors. Protease inhibitors appear to increase the release of fat from liver cells in vitro, at least in the presence of oleic acid (mimicking the intake of a fatty meal). In cultures of adipocytes, combining ritonavir with either AZT or d4T causes more release of fat from these fat cells than with RTV, AZT or d4T alone. Ritonavir, but not indinavir, leads to rises in cholesterol and triglycerides in healthy volunteers within a matter of weeks of starting therapy. Clinical studies suggest lopinavir/ritonavir is associated with more cholesterol and triglyceride rises than nelfinavir (in Abbott's M98-863 study), while the new protease inhibitor, atazanavir, has few, or perhaps no, effects upon lipids relative to nelfinavir (see ECCAT 2001 report). Previous studies with protease inhibitor-sparing regimens have suggested that they have more favorable (especially nevirapine in the Atlantic and Combine studies) or less unfavorable effects on lipids than protease inhibitor-based therapy. It has not yet been established whether or not rises in lipids shortly after the initiation of therapy are a harbinger of future lipodystrophy.

At the Athens lipodystrophy meeting, the ESS40002 study reported 24-week lipid data on treatment-naive patients randomized to a protease inhibitor-sparing regimen (AZT+3TC+ABC, n=86), or the protease inhibitor nelfinavir with either AZT+3TC (n=89) or d4T+3TC (n=83). Patients had similar baseline characteristics with CD4 cell counts of 328-359 cells/mm³ and viral loads of 4.0-4.8 log10.

At week 24, virological efficacy was similar across all groups. Cholesterol rose in all arms: 5.8, 21.6 and 29.3 mmol/l in the triple nucleoside, Nelfinavir+AZT+3TC and Nelfinavir+d4T+3TC arms respectively. HDL ("good") cholesterol levels rose 3.6, 2.6 and 3.8 mmol/l, respectively. It is not known why a relatively poorer rise in HDL was seen in the AZT+3TC+Nelfinavir arm -- for LDL levels fell marginally in the triple nucleoside arm and rose by approximately 14 and 22 mmol/l in the protease inhibitor-containing arms, respectively. Differences between males and females regarding changes in lipids were at best modest. Triglycerides rose in all arms: 7.7, 7.5 and 24.2 mmol/l, respectively. It is not known why a greater rise in triglycerides was seen in the d4T-based arm. Rises in CD4 counts favored the protease inhibitor-based therapies, with 90, 132 and 137 cells/mm³ rises observed, respectively over 24 weeks.

Regarding adverse events, diarrhea was seen in just 3% of triple nucleoside patients but 51 and 46% in the nelfinavir groups with AZT+3TC and d4T+3TC, respectively. On the other hand, nausea predominated in the AZT-containing arms with 29% on triple nucleoside and 31% on nelfinavir compared to 17% on d4T. Intriguingly, fatigue was reported in 13% of each of the AZT arms and just 4% of the d4T arms with hematological changes (anemia or low white cells) in 5%, 13% and 1% respectively.

The data support the premise that protease inhibitor-sparing regimens tend to be less lipid-unfriendly than regimens containing currently approved protease inhibitors. Differences in lipid changes as caused by AZT and d4T may exist, but the net effects on the total HDL:cholesterol ratio would be similar. Changes in lipids seem to be similar with regard to gender. In individuals at known risk for future cardiovascular disease (for example, male smokers over 45 with a family history), consideration should be given to using regimens that have the least unfavorable impact on lipids.