HIV Articles  
Switching to Atazanavir Improves Metabolic Disorders in Antiretroviral-Experienced Patients With Severe Hyperlipidemia  
  JAIDS Journal of Acquired Immune Deficiency Syndromes: Volume 39(2) 1 June 2005
Möbius, Ulrike MD*; Lubach-Ruitman, Margrit MSc*; Castro-Frenzel, Brigitte MD*; Stoll, Matthias MD*; Esser, Stefan MD†; Voigt, Esther MD‡; Christensen, Stefan MD§; Rump, Jörg-Andres MD‖; Fätkenheuer, Gerd MD¶; Behrens, Georg M. N MD*; Schmidt, Reinhold E MD*
From the *Department of Clinical Immunology, Hannover Medical School, Hannover, Germany; †Department of Dermatology, University Hospital Essen, Essen, Germany; ‡Department of Medicine, Rheinische Friedrich-Wilhelms-Universität Bonn, Bonn, Germany; §HIV specialist prac-tice, Münster, Germany; ‖HIV specialist practice, Freiburg, Germany; and ¶Department of Medicine I, University Hospital Köln, Köln, Germany.
The Early Access Program nested study was supported by Bristol-Myers Squibb, Germany; the Competence Network HIV/AIDS, C15; and the München H.W. & J. Hector Stiftung, Neinheim, Germany. Representatives of the pharmaceutical company had no role in gathering, analyzing, or interpreting the data.
Objective: To describe the efficacy and change in lipid profile in patients with severe hyperlipidemia after switch to an atazanavir-containing highly active antiretroviral therapy regimen.
Design and Methods: Open-field, 24-week, prospective observational cohort study including 33 HIV-infected, antiretroviral-experienced patients with hyperlipidemia. Changes in lipid profiles were evaluated by analyses of triglycerides, total cholesterol, high- and low-density lipoprotein (HDL and LDL) cholesterol, and efficacy by HIV RNA and CD4 cell changes, both from baseline to week 24.
Results: A rapid and significant decrease of 46% (5.81 ± 4 mmol/L vs. 3.16 ± 2.6 mmol/L, P = 0.002) in triglyceride levels was shown. Similarly, a sustained improvement of 18% was observed in total cholesterol levels during the first 24 weeks after switching to atazanavir (6.45 ± 1.9 mmol/L vs. 5.3 ± 1.3 mmol/L, P = 0.001). After 24 weeks of treatment there was a significant decrease of 22% in non-HDL cholesterol (5.76 ± 1.9 mmol/L at baseline vs. 4.5 ± 1.3 mmol/L at 24 weeks; P = 0.003). HDL and LDL cholesterol profiles did not change significantly as did the viral load or CD4 cell count.
Lipid Profiles
At 24 weeks, there was a significant decrease of 45.6% in triglycerides (5.81 ± 4 mmol/L at baseline vs. 3.16 ± 2.6 mmol/L at 24 weeks; P = 0.002) and of 17.9% in total cholesterol (6.45 ± 1.9 mmol/L at baseline vs. 5.3 ± 1.3 mmol/L at 24 weeks; P = 0.001) referring to all patients. Considering only the fasting measures, a significant decrease of 50% in triglycerides (6.65 ± 4.5 mmol/L at baseline vs. 3.33 ± 3 mmol/L at 24 weeks; P = 0.006) and of 20% in total cholesterol (6.49 ± 2 mmol/L at baseline vs. 5.21 ± 1.4 mmol/L at 24 weeks; P = 0.01) could be shown. After 24 weeks of treatment there was a significant decrease of 21.7% in non-HDL cholesterol (5.76 ± 1.9 mmol/L at baseline vs. 4.5 ± 1.3 mmol/L at 24 weeks; P = 0.003) in the whole study population; fasting measures showed a 23% reduction in non-HDL cholesterol (5.81 ± 2.1 mmol/L at baseline vs. 4.48 ± 1.4 mmol/L at 24 weeks; P = 0.025). The changes in concentration of total cholesterol from baseline through 24 weeks are shown in Figure 2A-C. Both fasting and nonfasting conditions showed significant changes in triglyceride and cholesterol concentrations after 24 weeks of treatment. Compared with total cholesterol and triglycerides, changes in LDL and HDL cholesterol concentrations were not statistically significant in fasting (not shown) and nonfasting conditions. In contrast to the reduction of triglyceride levels until week 24, a significant re-increase in triglyceride levels could be detected after 48 weeks (week 24: 2.31 ± 1.1 mmol/L vs. 3.41 ± 1.7 mmol/L at week 48; [n = 14]; P = 0.045), although these levels were still lower than compared with baseline (6.72 ± 4.8 mmol/L). Total cholesterol levels remained reduced as compared with baseline after 48 weeks of treatment (7.33 ± 1.7 mmol/L at baseline vs. 5.43 ± 1.0 mmol/L at 48 weeks; P = 0.001). Coexistent with the increase in triglycerides was a relative increase in the proportion of patients receiving ritonavir-boosted atazanavir: due to a switch from atazanavir 400 mg to ritonavir-boosted atazanavir in week 24 (7 of 20 patients), the total number of patients receiving ritonavir-boosted atazanavir increased to 18 patients (from 10 initially). Therefore, the patients on ritonavir-boosted atazanavir represented almost 60% (9 of 16 patients) of the remaining study population from week 24 on.
Conclusions: Switching to atazanavir results in a significant improvement in HIV therapy-induced hyperlipidemia. A switch to atazanavir is proposed as a valuable option to improve atherogenic lipid profiles while maintaining virologic control.
Virologic and Immunologic Outcome and Treatment Failure
At baseline only 16 (48.5%) of all 33 participants had HIV-1 RNA levels <50 copies/mL. Overall, 19 of 33 study subjects (57.6%) showed HIV-1 RNA levels <50 copies/mL after 24 weeks of treatment and 9 of 16 (56.3%) after 48 weeks. Mean HIV RNA and CD4 cell count changes from baseline to 24 or 48 weeks were not statistically significant. Two patients (6.1%) experienced virologic failure under atazanavir therapy defined as increase in HIV RNA levels above the level of quantification (50 copies/mL) from baseline and 10 patients (30.3%) remained having detectable viral replication. All patients with detectable viral replication had been heavily pretreated prior to starting the study. There was no limitation to the design of the backbone therapy during the study. The backbones used are listed in Table 2.
This study demonstrates the beneficial effects on hyperlipidemia after switching to an atazanavir-containing regimen in pretreated HIV-infected patients. Patients receiving atazanavir therapy showed a significant decrease in triglyceride and total cholesterol levels. Most of these changes occurred in as little as 4 weeks after switching the regimen. Reduction of triglycerides was more pronounced than reduction of total cholesterol and remained stable during the total study period of 24 weeks. Such improvement may have an influence in reducing the risk of cardiovascular effects, the need for lipid-lowering agents, and the associated risks of drug-drug interactions. In addition, reduction in high triglyceride levels may lead to a significant decrease in the risk for developing acute pancreatitis. Finally, hypertriglyceridemia and increased free fatty acids have been proposed to significantly contribute to insulin resistance in HIV-associated lipodystrophy together with the direct drug-mediated mechanisms.22 We therefore speculate that a beneficial intervention on these factors could potentially also improve fat redistribution, peripheral and hepatic insulin resistance, and hepatic steatosis.
Although only incompletely assessed in our study, many HIV-infected patients receiving PI therapy develop signs and symptoms of the metabolic syndrome including hypertriglyceridemia, hypercholesterinemia, low HDL cholesterol, insulin resistance, and central obesity. The clinical features of the metabolic syndrome in HIV-seronegative patients are associated with increased risk of cardiovascular disease, including greater risk of CHD for a given level of LDL cholesterol, and premature death. Several studies indicate the positive association between the metabolic syndrome and cardiac morbidity. An approximately 50% greater risk of major coronary events in patients with the metabolic syndrome has been described.23,24 In these patients the size and number of VLDL particles are important determinants of the cardiovascular risk associated with hypertriglyceridemia. Although lacking VLDL characteristics in our study, we consider the baseline lipid profiles as closely related to the situation of the metabolic syndrome. To identify subgroups of patients with metabolic syndrome who are at substantially elevated cardiovascular risk and require intensive medical intervention, non-HDL cholesterol has been proposed as an additional indicator.21 Non-HDL cholesterol is thought to better encapsulate total risk from the atherogenic lipoproteins in patients with the metabolic syndrome, in whom LDL cholesterol is characteristically normal and triglyceride concentrations are modestly elevated.23 In agreement with this suggestion, a cohort study showed a stronger correlation of coronary mortality for non-HDL cholesterol than for LDL cholesterol.25 In addition, non-HDL cholesterol has become a secondary target of therapy in patients with elevated triglycerides (200-499 mg/dL) according to NCEP guidelines.8,21 Accordingly, we have included the non-HDL cholesterol in our analyses. In our study we observed a 21.7% decrease in non-HDL cholesterol after 24 weeks of treatment with atazanavir. Thus, switching to atazanavir successfully ameliorates secondary targets for lipid therapy in patients with normal LDL cholesterol and moderate to high hypertriglyceridemia8-a condition present in a significant proportion of our study cohort. Future studies should analyze the impact of a switch to atazanavir on the proportion of patients that achieve their individual lipid goals.
Among other factors, high levels of LDL cholesterol and low levels of HDL cholesterol have been identified as risk factors for CHD in the general population.26,27 Given that the use of ART, including PIs, is often associated with increases in total and LDL cholesterol,5,28,29 this has led to concern about HIV-infected patients being at a higher risk of developing atherosclerosis30 and therefore CHD. As mentioned earlier, marked elevations of triglycerides are considered an independent CHD risk factor, suggesting that some triglyceride-rich lipoproteins are atherogenic, one of them being VLDL. As the lipid content of VLDL is 80% triglycerides and 20% cholesterol, the increased concentration of VLDL would lead to both elevated total cholesterol and increase in triglycerides. Given the improvement in total cholesterol and triglycerides together with almost normal LDL cholesterol at baseline (2.63 mmol/L), our findings suggest an increase in VLDL cholesterol as one cause for high total cholesterol levels in HIV-positive patients on ART.31,32
Former studies in therapy-experienced patients have reported favorable effects on lipid levels after switching to atazanavir therapy. The AI424-044 study revealed a 28% decrease in fasting triglycerides, a 16% reduction in total cholesterol, and a 21% decrease in fasting LDL cholesterol levels 12 weeks after switching from nelfinavir to atazanavir therapy.33 The AI424-009 study revealed a 27% decrease in fasting triglyceride levels under a dosage of atazanavir (600 mg) compared with a 6.7% reduction in LDL cholesterol after 48 weeks34; surprisingly ATV (400 mg) did not show a significant reduction in triglycerides or LDL cholesterol levels. In our present study, we are able to show a 45.6% reduction in triglyceride levels over 24 weeks and a 17.9% reduction in total cholesterol levels in selected patients with significant hyperlipidemia. Interestingly, there was no significant change in LDL or HDL cholesterol levels during the study period of 24 weeks.
Switch of antiretroviral drugs to treat hyperlipidemia and glucose intolerance has been evaluated in several studies. In most of them, PIs were replaced by either an NRTI or NNRTI, resulting in variable effects from improvement of metabolic parameters to no significant effect. Although many of these trials revealed a maintained viral suppression when NNRTIs were selected, concerns remained regarding the class switch within a potent regimen resulting in limited options to respond to viral mutations and viral rebound. In contrast, our study shows the metabolically effective switch of a regimen without introducing a new class of HIV drugs. It also provides additional evidence that metabolic abnormalities are not a class-specific effect of PIs.
The observation of increasing triglyceride levels at week 48 might be explained by an increased proportion of patients receiving ritonavir-boosted atazanavir regimens after week 24. It has been shown that other ritonavir-containing regimens are associated with pronounced elevations of triglyceride levels.35 Due to the small number of patients receiving ritonavir-boosted atazanavir at baseline being followed until week 48, our study is not powered to describe significant decreases in triglyceride changes in the group. Whether the significant improvement of lipids is still significant using exclusively ritonavir-boosted atazanavir-containing regimens needs to be assessed in further prospective studies with larger cohort samples. This is of clinical relevance as the current approval for atazanavir in Europe is restricted to pretreated patients and to the boosted regimen.
Limitations of our study should be considered for appropriate interpretation. First, the relatively low case numbers in recorded HDL and LDL cholesterol could contribute to the lack of significance. Larger cohort samples will be needed to further evaluate this. Second, a long-term follow-up is needed to demonstrate whether the changes described will translate into a clinical benefit, ie, further normalization of lipid levels under atazanavir therapy and maintained viral suppression. In addition, due to a lack of information on pretreatment lipid levels of the included patients, it is impossible to exclude the possibility that dyslipidemia occurred before exposure to HAART. This, however, would even underestimate the favorable effects of atazanavir as we would not expect a direct lipid-lowering effect of atazanavir. Finally, the study population included in the analyses was heterogeneous in terms of patients' treatment histories. It is possible that the NRTI backbone contributed to the differences in lipid profiles, given that an improvement in lipids has been described in patients switching from stavudine to tenofovir.36 Conversely, the study populations and treatment history could be more representative for the daily clinical situation seen by HIV physicians.
In conclusion, we are able to show that switching patients with hyperlipidemia to an ART containing atazanavir leads to a significant reduction in total cholesterol, non-HDL cholesterol, and triglyceride levels and is therefore a treatment option in patients with a high-risk lipid profile for cardiovascular events. In addition to the favorable lipid changes, therapy with atazanavir appears to be safe, effective, and well tolerated. It remains to be addressed to what extent the changes in the lipid profile correlate with a decrease in cardiovascular events.
Adverse Events and Treatment Discontinuation
Two of 33 patients (6.1%) discontinued the study prior to week 24: 1 subject developed abdominal pain and did not continue scheduled visits and 1 subject discontinued treatment because of virologic failure. Six patients (18.2%) developed adverse events. Side effects included icterus in 2 patients and gastrointestinal symptoms (nausea, flatulence, abdominal pain) in 4 patients (12.1%). One patient additionally developed a rash, which was possibly related to atazanavir and 1 patient reported increased erectile dysfunction. All reported adverse events were designated as nonserious and were classified as mild to moderate (grade 1 or 2).
Highly active antiretroviral therapy (HAART) has led to a significant decrease in morbidity and mortality and to a notable extension of life expectancy in HIV-infected patients.1 The benefits of antiretroviral combinations, however, are tempered by a broad spectrum of side effects, including metabolic disorders such as hyperlipidemia, hyperglycemia, insulin resistance, and lipodystrophy. These metabolic and clinical changes resembling features of the metabolic syndrome have been reported during the course of HIV infection and its treatment with HAART, especially in protease inhibitor (PI)-based regimens.2,3 In particular, dyslipidemia occurs in up to 70%-80% of HIV-infected individuals receiving HAART and has been associated with almost all of the available PIs.4 Dyslipidemia implies significantly higher fasting total cholesterol, triglycerides, low-density lipoprotein (LDL) cholesterol, and very low-density lipoprotein (VLDL) cholesterol, with hypertriglyceridemia being the major lipid alteration.3,5,6
Hypertriglyceridemia and elevated total and LDL cholesterol levels, together with low high-density lipoprotein (HDL) cholesterol, are known to increase cardiovascular risk in the general population and may similarly predispose HIV-infected subjects to accelerated coronary illness.7-10 The D:A:D study (Data Collection on Adverse Events of Anti-HIV Drugs) recently reported a 26% increase in the estimated risk of myocardial infarction per year of exposure to PI-containing combination antiretroviral therapy (ART).11
Switching from a PI-based treatment to a PI-sparing regimen has been evaluated. Several studies have demonstrated that replacing a PI-containing regimen with nevirapine, efavirenz, or abacavir results in a significant improvement in dyslipidemia.12-14 Compared with treatment with nevirapine, the use of efavirenz was associated with higher total cholesterol and triglyceride levels. HDL cholesterol levels, however, were similar in patients receiving either efavirenz- or nevirapine-containing regimens.11 There are data as well suggesting that the extent of the described metabolic disturbances could differ according to different drugs within the PI class itself. Ritonavir-containing regimens were associated with the most pronounced elevations of total cholesterol and triglyceride levels15-17; saquinavir-containing regimens were associated with a lower risk of having an elevated ratio of total cholesterol/HDL cholesterol11; and nelfinavir-containing regimens were associated with a lower risk of having a reduced HDL cholesterol level.18 Fosamprenavir, as another new PI, seems to have a similar favorable lipid profile when compared with nelfinavir as shown in the SOLO study (a 48-week efficacy and safety comparison of once-daily fosamprenavir/ritonavir versus twice-daily nelfinavir in naive HIV-1 infected patients).19
Given the current need for lifelong therapy, considerations of long-term toxicities such as hyperlipidemia and lipodystrophy are becoming increasingly important when choosing between different regimens. New drugs are more likely to find a place in the management of HIV infection if they have high bioavailability, a distinct resistance profile, greater potency, a favorable pattern of adverse effects, and a lower rate of long-term complications. Atazanavir is a new potent and safe azapeptide PI with a high inhibitory quotient and a pharmacokinetic profile that allows once-daily oral administration. In contrast to other PIs, atazanavir is not supposed to be associated with increases in total cholesterol, LDL cholesterol, or triglycerides.20
The objective of this clinical study was to describe the changes in lipid profiles and efficacy of atazanavir-containing regimens in patients with severe hyperlipidemia. We aimed to investigate whether atazanavir is an option for patients with hyperlipidemia.
Baseline Characteristics
A total of 33 HIV-infected patients with severe hyperlipidemia were included in the study. Table 1 summarizes the baseline characteristics for all subjects. The total study population was 94% male and 6% female and the mean age was 46 years. Median HIV RNA at baseline was 72 copies/mL and mean CD4 cell count was 455 cells/μL. A total of 26 patients (78.8%) had a PI-containing regimen before switching to the atazanavir-containing therapy; 9 (27.3%) were on a PI-sparing regimen prior to inclusion into the study. At baseline 21 of all study subjects (63.6%) received 400 mg of atazanavir, and 10 subjects received 300 mg of atazanavir together with 100 mg of ritonavir (30.3%). There was no significant difference between mean fasting and nonfasting lipid values at baseline.
Mean TG: 5.9 mmol/L
Total chol: 6.55 mmol/L
LDL chol: 2.79 mmol/L
HDL chol: 0.9 mmol/L
Mom HDL chol: 5.76 mmol/L
Fasting TG: 6.67 mmol/L
Fasting total chol: 6.62 mmol/L
Fasting LDL chol: 2.63 mmol/L
Fasting HDL: 0.87 mmol/L
Study Design
Study AI424-900 is an open-label, multicenter, noncomparative phase IIIb study in HIV-infected patients who did not tolerate ART or in whom ART had failed. The substudy described here is an observational, prospective, 24-week cohort study including HIV-infected, antiretroviral-pretreated patients with severe hyperlipidemia. The primary objective was to investigate the changes in lipid profiles in these patients. Secondary objectives were the investigation of safety, tolerability, and antiretroviral efficacy after switch to atazanavir. Subjects either received atazanavir (400 mg, unboosted) once daily or atazanavir (300 mg) in combination with ritonavir (100 mg, boosted) once daily. The treating physician individually selected 2 or 3 nucleoside reverse transcriptase inhibitors (NRTIs) as a backbone. All drugs were used open label.
The study was conducted in accordance with the Declaration of Helsinki. All patients gave written informed consent before participating. Institutional review boards and independent ethics committees approved the informed consent and the protocol prior to initiation of the trial and all protocol amendments.
Study Population
Antiretroviral-experienced HIV-1-infected patients were eligible for inclusion if they were at least 16 years of age and had severe hyperlipidemia under the current ART. The threshold levels for triglycerides (≥2 mmol/L), total cholesterol (≥5 mmol/L), HDL cholesterol (≤1 mmol/L), and LDL cholesterol (≥3 mmol/L) were based on cutoff values for a higher risk of coronary heart disease (CHD), as outlined in the US National Cholesterol Education Program (NCEP) guidelines.21 Additional inclusion criteria were treatment failure defined as ART resistance, other intolerance, or adherence problems with inability to construct an alternative effective HAART regimen using alternative available ART agents. None of the patients received lipid-lowering therapy.
Study Assessments
Subject evaluation was performed at screening; baseline; weeks 4, 8, 12, and 24; and weeks 36 and 48 (follow-up visits). The clinical laboratory testing was performed in local laboratories used by the investigative sites. Laboratory tests included HIV RNA, absolute CD4 cell count, lipid profile (including total cholesterol, LDL and HDL cholesterol, and triglycerides), bilirubin, alanine aminotransferase, and aspartate aminotransferase, pancreatic lipase, glucose, creatinine, hemoglobin, and leukocyte and platelet counts. Triglycerides, total cholesterol, and HDL cholesterol were measured directly in local laboratories. LDL cholesterol was measured directly in serum samples of 27 patients; in 6 patients the Friedewald method for calculating LDL cholesterol values was used provided that triglyceride values were <4.45 mmol/L. In accordance with the NCEP guidelines, the non-HDL cholesterol, defined as total cholesterol minus HDL cholesterol, was additionally analyzed. The safety of the study regimens was assessed using patients' medical histories, physical examination, clinical laboratory test results, and reporting of adverse events, grouped using a modified version of the FDA Coding Symbols for Thesaurus of Adverse Reaction Terms (COSTART). Severity was evaluated according to the modified World Health Organization (WHO) criteria and rated on a scale of 1-4. Dose modification was permitted if toxicities were encountered.
Statistical Analyses
All analyses were exploratory. Differences between week 4, 8, 12, 24, 48, and baseline values were calculated for lipid values. The statistical significance of the longitudinal changes in these parameters and CD4 cell count was assessed using the paired Student t test or Bonferroni test for general linear models. All comparisons used a 2-sided α level of 0.05. Changes in plasma HIV RNA level were calculated using the Wilcoxon test.
  icon paper stack View Older Articles   Back to Top