icon star paper   Articles  
Back grey_arrow_rt.gif
Reyataz vs Nelfinavir: phase II study
  Dose-ranging, randomized, clinical trial of atazanavir with lamivudine and stavudine in antiretroviral-naive subjects: 48-week results
AIDS 2003; 17(18):2603-2614
Robert L. Murphy; Ian Sanne a; Pedro Cahn b; Praphan Phanuphak c; Lisa Percival d; Thomas Kelleher d; Michael Giordano d
Objective: To compare the efficiency and safety of atazanavir and nelfinavir in antiretroviral-naive patients.
Design: Randomization to atazanavir 400 mg or 600 mg once daily; nelfinavir 1250 mg twice a day, plus lamivudine and stavudine.
Methods: A blinded (to the atazanavir dose), 48-week trial in patients with HIV-1 RNA >2000 copies/ml, CD4 cell count >100 cells/l. Primary end-point: change in HIV-1 RNA from baseline at 48 weeks. Secondary end-point: subjects with HIV-1 RNA < 400, and < 50 copies/ml, CD4 cell count changes, adverse events.
Results: The 467 randomized subjects had comparable baseline characteristics across treatments. With atazanavir 400 mg, 600 mg and nelfinavir, respectively, mean changes in HIV-1 RNA (log10 copies/ml) from baseline to 48 weeks were -2.51, -2.58, -2.31; HIV-1 RNA < 400 copies/ml [intent-to-treat population (ITT), non-completion = failure (NC = F)], 64% (ATV 400), 67% (ATV 600), 53% (NFV); HIV-1 RNA < 50 copies/ml (ITT NC = F), 35%, 36%, 34%; mean CD4 cell count increased comparably at 48 weeks (234, 243, 211 cells/l).
Adverse events were similar across treatments with the exception of diarrhea (more frequent with nelfinavir) and jaundice (more frequent with atazanavir). Mean changes from baseline to 48 weeks were: fasting low density lipoprotein (LDL) bad cholesterol, +5.2%, +7.1% and +23.2% (at 56 weeks) and fasting triglycerides (48 weeks), +7.2%, +7.6% and +49.5%, in the atazanavir 400 mg, 600 mg, and nelfinavir groups, respectively (P < 0.01, atazanavir versus nelfinavir).
Conclusions: Atazanavir is a potent, safe, well tolerated, and effective once-daily protease inhibitor with low pill burden (two capsules/day). Lipid changes with atazanavir were significantly less than with nelfinavir, however, clinical significance of these finding in terms of decreased cardiovascular risk is unknown.
Highly active antiretroviral therapy (HAART) has proved effective in suppressing HIV-1 replication and improving clinical outcome. Low potency, variable pharmacokinetic characteristics, heavy pill burden, complex dosing requirements and metabolic toxicities, however, often limit the effectiveness of currently available protease inhibitor (PI)-based regimens, contributing to poor tolerability, suboptimal adherence and treatment failure. Currently available PIs cause potentially serious metabolic complications, significantly elevating lipids and triglycerides possibly contributing to lipodystrophy and increasing cardiovascular disease risk. The Adult AIDS Clinical Trials Group recommends that PI-related dyslipidemia be treated according to general population guidelines. However, lipid-lowering agents may introduce drug-related complications, as current PIs alter the metabolism of many of these agents, and their use potentially introduces additional complexity to already complicated HAART regimens. There is a need for effective PIs with simplified dosing without elevations in lipid concentrations.
Atazanavir is a PI in clinical development with an in vitro 50% inhibitory concentration of 2.6-5.3 nM and a distinct resistance profile. Its minimum serum concentration with sequential once-daily 400 mg dosing is greater than the protein binding-adjusted 50% inhibitory concentration for more than 36 h at steady state. It has a low pill burden (two capsules/dose). Its inhibitory quotient range is higher than that of current PIs (10.2 to 25.5). In vitro, atazanavir-resistant variants emerge more slowly than do variants resistant to either ritonavir or nelfinavir, and variants of HIV-1 that are resistant to up to three current PIs retain sensitivity to atazanavir. Atazanavir-resistant variants arising in vivo have a unique mutational pattern, including an I50L signature substitution. In an otherwise wild-type HIV-1 protease, the I50L substitution confers increased susceptibility of HIV-1 to other PIs, including amprenavir, but the clinical significance of this in vitro observation is not known. No cross-resistance has been observed between atazanavir and amprenavir in vitro in any of the isolates containing the I50L or I50V substitutions.
Comparative clinical trials show that atazanavir is safe, well tolerated and effective in rapidly and durably suppressing HIV-1 RNA and durably increasing CD4 cell count in both antiretroviral-naive and -experienced subjects for 48 weeks. Changes from baseline in lipid concentrations with atazanavir were not of a clinically significant magnitude, compared with the prompt, marked and sustained lipid elevations associated with comparator drugs. Based on safety and antiviral efficacy considerations, the 400 mg dose of atazanavir once daily was chosen for evaluation in Phase III studies.
The objective of this clinical trial, Study AI424-008, was to compare the safety and efficacy of atazanavir, once daily at two different doses, with twice-daily nelfinavir, each administered with stavudine and lamivudine. Preliminary results from the present trial have been presented previously.
The reasons for discontinuation were comparable across treatment groups. There were 54 (12%) discontinuations prior to week 48, 19 (4%) due to adverse events that were also comparable across treatment groups. Demographic characteristics were comparable across regimens. Mean HIV-1 RNA level was 4.74 log10 copies/ml, and the mean CD4 cell count was 295 cells/l.
The primary end-point was week 48. At week 24, mean changes from baseline in both the atazanavir 400 mg and 600 mg groups were -2.54 (0.07) log10 copies/ml, and in the nelfinavir group they were -2.42 (0.09) log10 copies/ml. These changes were sustained at 48 weeks.
The incidence of adverse events was comparable across regimens with the exception of diarrhea and jaundice. Diarrhea occurred significantly more frequently in the nelfinavir group than in either of the atazanavir groups (P < 0.0001). Jaundice occurred only with atazanavir, at an incidence of 11% in the 400-mg group and 20% in the 600-mg group (P < 0.0001, each atazanavir regimen versus nelfinavir). Scleral icterus occurred only with atazanavir treatment, at an incidence of 9% in the 400-mg group (P = 0.002, atazanavir 400 mg versus nelfinavir) and of 12% in the 600-mg group (P = 0.0001, atazanavir versus nelfinavir). Lipodystrophy as a clinical adverse event, as assessed by the investigator, was reported in seven subjects (4%) in the atazanavir 400-mg group, seven subjects (4%) in the 600-mg group and in two (2%) in the nelfinavir group. Although there is currently no case definition for lipodystrophy, all but one of these events were reported as grade 1 or 2. The one exception, grade 3 to 4, occurred in a patient treated with atazanavir 600 mg.
A total of 27 (6%) subjects discontinued due to one or more adverse events: nine subjects (5%) in the atazanavir 400-mg group, 14 (7%) in the atazanavir 600-mg group and four (4%) in the nelfinavir group. These 27 subjects reported 47 adverse events as leading to discontinuation, and 39 were considered at least possibly related to the study drug. Four atazanavir subjects (two in the 400-mg group and two in the 600-mg group) discontinued due to lactic acidosis or symptomatic hyperlactatemia (SHL) ranging from grade 1 to 3. Four atazanavir subjects in the 600-mg group discontinued due to hyperbilirubinemia that was attributed to the study drug. Two atazanavir subjects (one in each dose group) discontinued as a result of lipodystrophy that was attributed to the study drug. There were three deaths in the study, two in the atazanavir 600-mg group due to lactic acidosis on days 276 and 347, and one in the atazanavir 400-mg group from suicide at day 77.
Grade 3 to 4 elevations in total bilirubin were dose related and occurred most frequently in the atazanavir-treated subjects: 41% (400-mg group) and 58% (600-mg group), versus 4% in the nelfinavir group. Bilirubin elevations were reversible and predominantly indirect and unconjugated. Grade 3 to 4 elevations in AST and ALT were infrequent across treatment groups, and, as a marker of hepatotoxicity, showed no correlation with grade 3 to 4 elevations in total and indirect bilirubin.
Lactic acidosis syndrome or SHL developed in seven subjects, three in the atazanavir 400-mg group and four in the atazanavir 600-mg group, for 19.6 cases per 1000 subject years; two of these died (both in the atazanavir 600-mg group). Lactic acidosis was reported by the investigators based on the presence of elevated lactate, presence of metabolic acidosis and typical symptoms. Several lactic acidosis cases presented as part of a syndrome of other mitochondrial-related disorders, including pancreatitis and hepatotoxicity. Lactic acidosis was reported in subjects in Europe (two), South America (two), Asia (two) and Africa (one). All lactic acidosis cases were female, and all but one was overweight (body mass index > 25 kg/m2) or obese (body mass index > 30 kg/m3). Lactic acidosis tended to occur late in the course of the study (> 30 weeks). A multivariate analysis did not identify increased risk of lactic acid syndrome/SHL for the atazanavir 400 mg (P = 0.55), atazanavir 600 mg (P = 0.31) or with the combined atazanavir groups (P = 0.35) compared to the nelfinavir group.
Changes in lipid profiles
Significant differences were observed in percent changes from baseline in serum lipid profiles between atazanavir and nelfinavir. At 48 weeks, the mean percent increases from baseline in the atazanavir 400-mg, atazanavir 600-mg and nelfinavir groups, respectively, were as follows: total cholesterol, 5.1, 5.9, 24.6; fasting LDL cholesterol (values at 56 weeks), 5.2, 7.1, 23.2; fasting triglycerides, 7.2, 7.6, 49.5 (P < 0.01, all comparisons to baseline; P <0.01, atazanavir versus nelfinavir at week 48). Increases in high density lipoprotein (good) cholesterol were comparable across treatment groups.
At baseline, the proportions of subjects with total cholesterol values in the various categories defined by the National Cholesterol Education Program (NCEP), Adult Treatment Panel III (ATP III), guidelines were comparable across regimens. At baseline, 79-83% of subjects had total cholesterol levels classified as desirable (< 200 mg/dl). At week 48, 75% of subjects in the atazanavir 400-mg group had maintained desirable total cholesterol concentrations. In the nelfinavir group, the proportion with desirable total cholesterol decreased from 82% at baseline to 49% at 48 weeks.
Study AI424-008 demonstrates that a HAART regimen with atazanavir 400 mg or 600 mg once daily is comparable in safety and has similar antiviral efficacy to nelfinavir twice daily, each in combination with lamivudine and stavudine through 48 weeks. AI424-008 is one of the first trials using a once-daily PI with a low pill burden, of two capsules/day. As in other indications, the simplicity of this dosing schedule, with a low incidence of side effects and better tolerability, is expected to encourage adherence and improve virological outcomes, leading to more durable HIV-1 suppression and a delay in the emergence of drug-resistant variants.
Subjects randomized to an atazanavir arm experienced significantly better virologic outcome in some analyses. At 48 weeks, the decline in HIV-1 RNA level was similar between the atazanavir 400-mg group and the nelfinavir group, but, in the atazanavir 600-mg group it was significantly greater than that in the nelfinavir group (P < 0.025). The proportions of virological responders were generally comparable across treatment groups. However, in the ITT analysis at 48 weeks, a significantly greater proportion in the atazanavir 600-mg group than in the nelfinavir group achieved HIV-1 RNA levels < 400 copies/ml (P < 0.05). In the on-treatment analysis, the proportion of responders at HIV-1 RNA < 400 copies/ml was also significantly greater in the atazanavir 400-mg and 600-mg treatment groups than in the nelfinavir treatment group at 48 weeks (P < 0.05, both atazanavir groups versus nelfinavir). Durable increases in CD4 cell count from baseline were comparable in the three treatment groups at 48 weeks. Efficacy results for atazanavir confirm the findings from previous trials in antiretroviral-naive subjects (trial AI424-007) and in antiretroviral-experienced subjects (trial AI424-009).
Atazanavir was well tolerated at both doses, and its safety was comparable to that of nelfinavir. The incidence of diarrhea was greater in the nelfinavir group than in both atazanavir groups (P < 0.0001). Reversible, dose-related elevations of indirect, unconjugated bilirubin were reported with greater frequency with atazanavir, and although asymptomatic, resulted in treatment discontinuation in four subjects in the atazanavir 600 mg group. Elevated bilirubin was not associated with hepatotoxicity as assessed by grade 3 to 4 elevations of ALT and AST. In the presence of elevated bilirubin, measurement of serum aminotransferases increases the sensitivity of detecting the presence or absence of liver damage. In acute hepatocellular necrosis caused by drug-induced hepatitis, elevated bilirubin levels are accompanied by a significant elevation in levels of ALT and AST. Grade 3 to 4 elevations of ALT and AST with atazanavir treatment were infrequent and comparable to nelfinavir. Preclinical studies have shown that atazanavir-associated bilirubin elevations may be attributable to inhibition of uridine diphosphate glucuronosyltransferase 1A1, a mechanism similar to that described for the reversible elevations in bilirubin associated with Gilbert's syndrome, which are of little clinical significance. This is also the apparent mechanism for reversible bilirubin elevations reported with indinavir treatment. Although the discontinuation rate for adverse events was similar across all treatment groups, the incidence of jaundice was 11% and 20% in the atazanavir 400-mg and 600-mg treatment groups, respectively. This observation accounted, in part, for the decision to carry the 400-mg dose to phase III clinical trials.
Lactic acidosis developed in seven subjects receiving atazanavir and contributed to two deaths in the study. Lactic acidosis has been reported in patients receiving HAART, particularly in obese women, and is typically associated with the nucleoside components of the regimen. In this study, which involved a 4 : 1 randomization to atazanavir- compared to nelfinavir-based treatment and all patients received the same nucleoside components, there were no identified risk factors for development of lactic acidosis.
Dyslipidemia occurs with current PIs and is implicated in development of cardiovascular disease. Dyslipidemia may coincide with PI-associated fat redistribution. In contrast to the prompt, marked, sustained and potentially clinically relevant increases in serum lipid and triglyceride concentrations observed with nelfinavir, significantly smaller elevations in total cholesterol, fasting LDL cholesterol or fasting triglyceride concentrations above baseline occurred with atazanavir treatment through 48 weeks (P < 0.01). Relatively small increases in lipid parameters may actually represent a return to pre-HAART lipid levels as suggested in a recent report from the Multicenter AIDS Cohort Study.
Current guidelines recommend that PI-associated dyslipidemia be treated to reduce the risk of cardiovascular events. For the general population, the NCEP ATP III classifies total cholesterol of < 200 mg/dl as desirable, 200-239 mg/dl as borderline high, and >240 mg/dl as high. Total cholesterol levels within the range of borderline high to high values can develop in HIV-1-infected patients treated with current PIs. In the present study, 17-21% of subjects had borderline high or high total cholesterol values at baseline. At weeks 24 and 48, 40% and 51% had developed borderline high or high total cholesterol concentrations, respectively, in the nelfinavir group, compared with 21% and 28% in the atazanavir groups. The Adult AIDS Clinical Trials Group recommends that PI-associated dyslipidemia be treated according to the NCEP recommendations, stressing the paramount importance of avoiding drug-drug interactions. The 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitors (the 'statins') are the most commonly used agents for such treatment. Significant drug interactions arise when PIs and statins are co-administered, as many statins are metabolized by the cytochrome P450 isoenzyme CYP 3A4, which the PIs inhibit. Ritonavir and saquinavir co-administered with atorvastatin or simvastatin increase their serum concentrations 31-fold, and similar elevations have been reported with nelfinavir and lopinavir/ritonavir. Elevated levels of statins have been associated with an array of toxicities, including rhabdomyolysis.
In the present study, the increases in serum lipids that occurred in the nelfinavir treatment arm were similar in magnitude to those that have been observed with many of the current PIs. Although the lipid changes induced by PIs and other agents used to treat HIV may have clinical consequences in the HIV population, the clinical significance of these findings in terms of increased cardiovascular risk is not known. The need to treat dyslipidemia introduces additional complexity and pill burden to already complex HAART regimens, potentially further reducing adherence. Less-than-perfect adherence leads to the emergence of drug-resistant HIV-1 variants. Boosting of a primary PI with ritonavir can reduce the number of required daily doses, but it also increases the total pill burden and the severity of metabolic side effects such as dyslipidemia, as well as introducing new side effects. Atazanavir offers an alternative strategy to ritonavir boosting, since it is effective taken once daily, with a low pill burden (two capsules/day) and the changes in lipids and triglycerides associated with its use are significantly less than those observed with other PIs, including nelfinavir.
Although there was no case definition for a lipodystrophy endpoint, two subjects, one each in the atazanavir 400-mg and 600-mg treatment groups, were discontinued from treatment because of investigator-assessed lipodystrophy. Based on the low incidence of lipodystrophy in this study, blinded only as to atazanavir dose, conclusions concerning the two Pis and lipodystrophy could not be made.
In summary, this trial in antiretroviral-naive subjects demonstrates that atazanavir once daily has safety, tolerability and efficacy comparable to nelfinavir twice daily through 48 weeks. Atazanavir rapidly and durably suppresses HIV-1 RNA levels and durably increases CD4 cell count. Atazanavir was significantly less likely than nelfinavir to cause diarrhea. Reversible, unconjugated hyperbilirubinemia, not associated with hepatotoxicity, was the most frequently reported laboratory abnormality in atazanavir-treated subjects. Changes in lipid and triglyceride levels with atazanavir treatment are of a significantly smaller magnitude than the prompt, marked, and sustained elevations observed with nelfinavir.
  icon paper stack View Older Articles   Back to Top   www.natap.org