Once-Daily Atazanavir/Ritonavir Compared With Twice-Daily Lopinavir/Ritonavir, Each in Combination With Tenofovir and Emtricitabine, for Management of Antiretroviral-Naive HIV-1-Infected Patients: 96-Week Efficacy and Safety Results of the CASTLE Study
Download PDF here
JAIDS Journal of Acquired Immune Deficiency Syndromes: March 1 2010
Molina, Jean-Michel MD*; Andrade-Villanueva, Jaime MD; Echevarria, Juan MD; Chetchotisakd, Ploenchan MD; Corral, Jorge MD; David, Neal FCFP#; Moyle, Graeme MD**; Mancini, Marco BS; Percival, Lisa MS; Yang, Rong PhD; Wirtz, Victoria MS; Lataillade, Max DO, MPH; Absalon, Judith MD, MPH; McGrath, Donnie MD, MPH ; for the CASTLE Study Team
From the *Department of Infectious Diseases, Hospital Saint-Louis, AP-HP; University of Paris-Diderot, Paris, France; Unidad de NIH, Hospital Civil De Guadalajara, Guadalajara, Mexico; Departamento de Enfermedades Infecciosas, Tropicales y Dermatologicas, Hospital Nacional Cayetano Heredia, Lima, Peru; Department of Medicine, Khonkaen University, Khonkaen, Thailand; Hospital Interzonal Gral. De Agudos Oscar Alende, Buenos Aires, Argentina; #Brooklyn Medical Centre, Western Cape, South Africa; **Kobler Clinic, St Stephens Centre, Chelsea and Westminster Hospital, London, UK; and Research and Development, Bristol-Myers Squibb, Wallingford, CT.
Background: Once-daily atazanavir/ritonavir demonstrated similar antiviral efficacy to twice-daily lopinavir/ritonavir over 48 weeks, with less gastrointestinal disturbance and a better lipid profile, in treatment-naive patients.
Methods: International, multicenter, open-label, 96-week noninferiority randomized trial of atazanavir/ritonavir 300/100 mg once daily vs lopinavir/ritonavir 400/100 mg twice daily, each in combination with fixed-dose tenofovir/emtricitabine 300/200 mg once daily, in antiretroviral-naive, HIV-1-infected patients. The primary end point was the proportion of patients with HIV RNA <50 copies/mL at 48 weeks. Results through 96 weeks are reported.
Results: Of 883 patients enrolled, 440 were randomized to atazanavir/ritonavir and 443 to lopinavir/ritonavir. At week 96, more patients receiving atazanavir/ritonavir achieved HIV RNA <50 copies/mL (74% vs 68%, P < 0.05) in the intent-to-treat analysis. On both regimens, 7% of subjects were virologic failures by 96 weeks. Bilirubin-associated disorders were greater in patients taking atazanavir/ritonavir. Treatment-related gastrointestinal adverse events were greater in patients taking lopinavir/ritonavir. Mean changes from baseline in fasting total cholesterol, non-high-density lipoprotein cholesterol, and triglycerides at week 96 were significantly higher with lopinavir/ritonavir (P < 0.0001).
Conclusions: Noninferiority of atazanavir/ritonavir to lopinavir/ritonavir was confirmed at 96 weeks. Atazanavir/ritonavir had a better lipid profile and fewer gastrointestinal adverse events than lopinavir/ritonavir.
The reductions in morbidity and mortality associated with combination highly active antiretroviral therapy (HAART) for HIV-1 infection in the developed world1 indicate that components of these regimens will continue to be used for increasingly longer-term treatment of patients. Globally, recommendations for the treatment of HAART-naive HIV-infected patients include regimens containing a ritonavir-boosted protease inhibitor (PI) or a nonnucleoside reverse transcriptase inhibitor combined with 2 nucleoside reverse transcriptase inhibitors (NRTIs).2-5 Established efficacy and highgenetic barrier to the emergence of resistance are key factors in the widespread inclusion of PIs in combination HAART regimens,6 with most international guidelines recommending the use of atazanavir, darunavir, fosamprenavir, lopinavir, or saquinavir as the PIs of choice.2-5
Published reports of comparative trials of ritonavir-boosted PIs have demonstrated similar efficacy between agents over 48 weeks in antiretroviral-naive patients infected with HIV-1, including fosamprenavir and lopinavir,7 darunavir and lopinavir,8 atazanavir and lopinavir,9 and saquinavir and lopinavir.10 Efficacy profile alone is therefore not a key differentiating drug-related factor influencing choice of therapy-tolerability and toxicity are also important considerations that help to differentiate between treatment options.11-14 In addition to efficacy and safety, factors such as pill burden and dosing frequency, which are known to impact adherence, and in turn treatment outcomes, need to be considered.11
The majority of clinical trials of HAART regimens report data from 48 weeks as the primary end point; however, data beyond the first year of therapy are important in differentiating ritonavir-boosted PI-containing HAART regimens with regard to continued efficacy, tolerability, and quality of life. The aim of the 96-week CASTLE study (BMS AI424138) was to examine the long-term comparative clinical efficacy and safety of once-daily atazanavir/ritonavir and twice-daily lopinavir/ritonavir, both given in combination with once-daily, fixed-dose tenofovir and emtricitabine, for the management of HIV-1 infection in antiretroviral treatment-naive patients through 96 weeks of therapy.
A detailed description of the study methods has been published and a brief description of the 96-week methodology is provided below.9
Patients were eligible for this study if they were aged 18 years or greater, infected with HIV-1, naive to antiretroviral therapy (less than 1 week of previous antiretroviral exposure, except in the setting of postexposure prophylaxis or prevention of mother-to-child transmission, in which case less than 6 weeks of previous antiretroviral exposure was allowed), and had HIV-1 RNA of 5000 copies/mL or greater. Patients were recruited by 134 centers in 29 countries between November 2005 and June 2006.
This study was performed in accordance with Good Clinical Practice and the ethical principles of the Declaration of Helsinki. The protocol was approved by the institutional review board at each study site, and patients provided written-informed consent before participation in the study.
The primary end point of the study was the proportion of patients with HIV RNA <50 copies/mL at week 48, and has previously been reported.9 Secondary efficacy end points were the proportion of patients with HIV RNA <50 copies/mL at week 96, the proportion of patients with HIV RNA <400 copies/mL at weeks 48 and 96, change in CD4 cell count from baseline through week 96, log reduction in HIV RNA by week 96, and the antiretroviral resistance profiles of patients experiencing virologic failure. Adherence to study regimen to week 96 was also assessed. Safety end points included the incidence of adverse events and serious adverse events, discontinuations due to adverse events, laboratory abnormalities, and changes from baseline in fasting lipids over time. Patients assigned to lopinavir/ritonavir were able to switch from the soft-gel capsule to the tablet formulation after 48 weeks in the event of intolerance to the soft-gel capsule formulation or for subject convenience.
Statistical methods for the week 48 analysis have been previously published.9 The week 96 analyses used the same methods and are briefly summarized below.
Efficacy results are presented by the as-randomized treatment regimen [intention to treat (ITT)]. Safety results are presented by the as-treated treatment regimen (ie, by the treatment regimen actually received). Two-sided tests of statistical significance at the 0.05 level were used. The proportion of subjects with HIV RNA <50 copies/mL at week 96 was assessed with several algorithms and cohorts of randomized subjects. The 96-week principal analysis was based on the confirmed virologic response (CVR) noncompleter = failure (NC = F; ITT) definition of response. Supportive analyses used the CVR noncompleter = missing (NC = M), time to loss of virologic response (TLOVR: an ITT analysis that defines response as 2 consecutive on-treatment HIV RNA <50 copies/mL achieved and maintained through week 96 without intervening discontinuation and virologic rebound), and virologic response-observed cases (VR-OC) definitions of response.
The treatment regimens were compared using the difference in proportions (atazanavir/ritonavir - lopinavir/ritonavir) and 95% confidence interval (CI) based on a stratified normal approximation. Analyses were stratified by the same strata as randomization-that is, HIV RNA level at enrollment and geographic region. The atazanavir/ritonavir regimen was deemed to be noninferior to the lopinavir/ritonavir regimen if the lower CI for the difference in proportions was greater than -10%. The proportion of subjects with HIV RNA <400 copies/mL at week 96 was assessed analogously to the proportion with HIV RNA <50 copies/mL.
Observed values were used to summarize HIV RNA log10 levels, CD4 cell counts, and their changes from baseline through week 96. Mean CD4 changes from baseline at week 96 were compared between treatment regimens with 95% CIs based on stratified normal approximations using observed values. Descriptive analyses were performed for patients with HIV RNA <50 copies/mL at week 96 with CVR (NC = F) by prespecified baseline subgroups.
Genotypic and phenotypic resistance profiles were determined for patients who met criteria for virologic failure through week 96 as defined by CVR (NC = F) for HIV RNA of 400 copies/mL or more. Virologic failure was defined as rebound after achieving a confirmed HIV RNA <400 copies/mL without resuppression, discontinuation due to insufficient HIV RNA response before week 96, or failure to achieve a confirmed HIV RNA of <400 copies/mL and on study at week 96.
The proportions of patients adherent to the regimens at week 96 were determined by comparing actual study medications received with those reported on the Multicenter AIDS Cohort Study (MACS) adherence questionnaire for treated patients with evaluable results.15,16 Patients were considered adherent to regimen only when they were adherent to all drugs in the regimen. Nonadherence to a drug resulted when 1 of the following criteria was met: (1) patient took fewer medications than prescribed in the past 4 days; (2) patient took medication as prescribed in the past 4 days in an atypical pattern; (3) patient took fewer pills per dose than prescribed in the past 4 days; or (4) patient provided partial responses to adherence questions.
Analyses of fasting lipids over time excluded values obtained after the start of serum lipid reduction therapy. Mean percent changes in fasting lipids from baseline were compared between treatment regimens with 95% CIs based on stratified normal approximations.
Role of the Funding Source
The study design and analysis plan were developed by the study sponsor with input from prospective investigators. Decisions regarding the final protocol, data reviews, and publishing were made based on discussion between the sponsor and the study investigators. The corresponding author and the sponsor had full access to the data after official closure of the database, and the corresponding author had final responsibility for submitting the manuscript.
Demographics and Disposition of Patients
The disposition of patients is shown in Figure 1. Of 883 patients included in the efficacy analysis, 440 were randomized to atazanavir/ritonavir and 443 were randomized to lopinavir/ritonavir. Baseline characteristics were similar between the treatment groups and have been published previously.9 Overall, 39 patients treated with lopinavir/ritonavir switched from the capsule to the tablet formulation between weeks 48 and 96. The mean/median time to tablet switch was 87 weeks. Most subjects who switched were men (N = 31) with the majority of subjects being from Asia (N = 14) and South America (N = 16).
Of the 438 patients who received study treatment in the atazanavir/ritonavir group, 72 (16%) discontinued before week 96 compared with 95 (22%) of the 440 patients who received study treatment in the lopinavir/ritonavir group; a greater proportion of these discontinuations in the lopinavir/ritonavir group was due to adverse events and withdrawal of consent (Fig. 1). Three (<1%) patients taking atazanavir/ritonavir discontinued due to jaundice/hyperbilirubinemia compared with none taking lopinavir/ritonavir. None of these discontinuations on atazanavir/ritonavir occurred between weeks 48 and 96. More patients taking lopinavir/ritonavir than atazanavir/ritonavir discontinued due to diarrhea [7 (1.6%) vs 0], with 2 discontinuations occurring between weeks 48 and 96. Renal adverse events led to discontinuation for 1 patient on each regimen, only during the first 48 weeks (Fanconi syndrome with atazanavir/ritonavir, proteinuria with lopinavir/ritonavir).
At week 48, the noninferiority of atazanavir/ritonavir to lopinavir/ritonavir was demonstrated, with similar proportions of patients in each group achieving the primary end point.9 At week 96, in the ITT analysis CVR (NC = F), 74% of patients in the atazanavir/ritonavir group achieved the end point of HIV RNA <50 copies/mL compared with 68% (P < 0.05) of patients in the lopinavir/ritonavir group (Table 1, Fig. 2). The atazanavir/ritonavir regimen met the criterion for noninferiority to the lopinavir/ritonavir-based regimen in this analysis and in the supportive efficacy analyses (CVR NC = M, TLOVR, VR-OC) (Table 1, Fig. 2).
The proportion of patients who achieved HIV RNA <400 copies/mL by the CVR (NC = F) analysis was 80% among patients taking atazanavir/ritonavir and 74% among patients taking lopinavir/ritonavir; the difference estimate was 5.1% (95% CI, -0.4% to 10.6%). Similar findings were observed in the additional analyses.
Responses by qualifying HIV RNA strata were higher for atazanavir/ritonavir than lopinavir/ritonavir for those patients with qualifying HIV RNA of <100 000 copies/mL and those with HIV RNA of 100,000 copies/mL or more (Fig. 3). There was no difference between groups in log reduction in HIV RNA from baseline to week 96 (mean HIV RNA reduction on atazanavir/lopinavir 3.21 log10 copies/mL vs 3.19 log10 copies/mL on lopinavir/ritonavir). Similar mean increases from baseline to week 96 in CD4 cell count were achieved with both regimens (268 cells/μL in the atazanavir/ritonavir group and 290 cells/μL in the lopinavir/ritonavir group; difference -21.2 cells/μL; 95% CI, -43.3 to 0.9).
Response rates (HIV RNA <50 copies/mL) were consistent on atazanavir/ritonavir across all the baseline CD4 subgroups, and were consistently higher than those achieved with lopinavir/ritonavir across all baseline CD4 subgroups (Fig. 4). In patients with baseline CD4 < 50 cells/μL, 78% responded in the atazanavir/ritonavir group compared with 58% in the lopinavir/ritonavir group (Table 2). Similarly, higher responses were demonstrated in patients with baseline HIV RNA 100,000 copies/mL or more and CD4 < 100 cells/μL randomized to atazanavir/ritonavir (71%) than lopinavir/ritonavir (61%) (Table 2).
Virologic failure [using CVR (NC = F)] occurred in similar numbers of patients in each group (Table 3). Overall, 11 patients taking atazanavir/ritonavir and 8 patients taking lopinavir/ritonavir who met criteria for virologic failure at week 48 were not virologic failures at week 96.
Among patients with virologic failure at week 48 or week 96 without baseline phenotypic resistance to their on-treatment PI, 3 developed major or minor PI substitutions through week 96. One patient in the atazanavir/ritonavir group had major and minor PI substitutions emerge and ultimately failed at week 67. This patient had baseline T12A/S, I13I/V, M36I, N37D, I62V, L63P, A71A/T, I72V, and I93L, with atazanavir fold-change of 0.78. In addition, 1 patient in the lopinavir/ritonavir group had minor PI substitutions emerge while on study. This patient had multiple major and minor substitutions (L10L/I, V32I, I54I/V, A71I, G73G/S and V82V/A, L89V, and L90M) without phenotypic resistance to lopinavir/ritonavir at baseline (LPV FC 6.09). An additional patient in the atazanavir/ritonavir group reported as a virologic failure at week 48 with an N88S substitution, associated with atazanavir resistance, subsequently achieved an HIV RNA <50 copies/mL at week 96 without a change in study regimen. No patient on either regimen with virologic failure that had a wild-type isolate at baseline developed genotypic or phenotypic resistance to atazanavir/ritonavir or lopinavir/ritonavir.
Patients in both the atazanavir/ritonavir and lopinavir/ritonavir treatment arms had similarly low rates of nonadherence (ie, <100% adherent) from week 4 to week 96 of the study. At study visits through week 96, across both treatment arms, 80% to 88% of patients were adherent to their treatment regimen (at week 96, adherence rates were 82% for atazanavir/ritonavir and 84% for lopinavir/ritonavir). For both regimens, the main reason for noncompletion of the MACS questionnaire was discontinuation of study medication [71/439 patients (16%) taking atazanavir/ritonavir and 90/433 patients (21%) taking lopinavir/ritonavir at week 96]. For patients who included reasons for nonadherence at week 96, the most common were "ran out of pills" in the atazanavir/ritonavir treatment group and "simply forgot" in the lopinavir/ritonavir treatment group.
The safety analysis was conducted on data from 878 treated patients (441 in the atazanavir/ritonavir group and 437 in the lopinavir/ritonavir group; 2 and 3 randomized patients on the atazanavir/ritonavir and lopinavir/ritonavir regimens, respectively, were never treated and 3 patients randomized to lopinavir/ritonavir received atazanavir/ritonavir). There were no unexpected safety events, adverse events were not treatment limiting in most cases, and the majority were mild to moderate in intensity.
There were no additional deaths after 48 weeks of the study. Serious adverse events were reported in 14% and 11% of patients on the atazanavir/ritonavir and lopinavir/ritonavir regimens, respectively, with an individual incidence of <1% in either regimen. Few were considered to be related to the study drugs.
The overall incidence of Grades 2-4 treatment-related adverse events was 30% and 32% in those patients on the atazanavir/ritonavir and lopinavir/ritonavir regimens, respectively (Table 4). Most of these were reported by 2% of patients or less on either regimen, and few were considered by the investigators to be related to the study drug (Table 4). More patients in the lopinavir/ritonavir treatment arm experienced gastrointestinal adverse events compared with the atazanavir/ritonavir arm. In addition, a larger proportion of subjects taking lopinavir/ritonavir (24%) compared with atazanavir/ritonavir (12%) required the use of antidiarrheal agents. Hepatobiliary adverse events were experienced by more patients in the atazanavir/ritonavir group than the lopinavir/ritonavir group. Three patients discontinued due to jaundice/hyperbilirubinemia through week 48 with no additional discontinuations due to hyperbilirubinemia occurring between weeks 48 and 96. Among patients who switched to the lopinavir/ritonavir tablet formulation (n = 39), the incidence of Grades 2-4 treatment-related adverse events after switch was 8%. There were no Grades 2-4 treatment-related adverse events of diarrhea after switch; diarrhea (all grades) was reported in 1/39 patients (3%) after switch.
The incidences of selected Grades 3-4 laboratory abnormalities are shown in Table 5. Elevations in total bilirubin were predictably higher in patients taking atazanavir/ritonavir than lopinavir/ritonavir (44% vs <1%). Grades 3-4 increases in alanine aminotransferase concentration were reported in 3% and 2% of patients taking atazanavir/ritonavir and lopinavir/ritonavir, respectively; Grades 3-4 elevations of aspartate aminotransferase were reported in 3% and 1% of patients taking atazanavir/ritonavir and lopinavir/ritonavir, respectively. Grades 3-4 elevations in serum creatinine were <1% in both regimens. Mean change from baseline to week 96 in serum creatinine was 0.04 mg/dL or less in both regimens, and substitution of tenofovir/emtricitabine for decline in renal function was <1% overall. The median calculated creatinine clearance percent change from baseline was -1% in patients in the atazanavir/ritonavir arm and -2% in the lopinavir/ritonavir arm. At week 96, the proportion of patients with >50% reduction from baseline in creatinine clearance was 0% in the atazanavir/ritonavir arm and <1% in the lopinavir/ritonavir arm.
Mean percent changes in lipids from baseline at week 96 were significantly higher (P < 0.0001) in patients taking lopinavir/ritonavir than atazanavir/ritonavir for fasting total cholesterol, non-high-density lipoprotein (HDL) cholesterol, and triglycerides (Fig. 5). Mean percent changes in low-density lipoprotein (LDL) cholesterol were similar between the 2 treatment groups. In addition, more patients on the atazanavir/ritonavir regimen than the lopinavir/ritonavir regimen had optimal lipids at week 96 as defined by the National Cholesterol Education Program Adult Treatment Panel III for total cholesterol, non-HDL cholesterol, LDL cholesterol, triglycerides, and total cholesterol:HDL cholesterol ratio.17 Shifts up of 1 category or more in NCEP for total cholesterol occurred in 16% of patients taking atazanavir compared with 29% taking lopinavir/ritonavir; 32% vs 40%, respectively, for LDL cholesterol; and 23% vs 49%, respectively, for triglycerides. A total cholesterol:HDL cholesterol ratio >5 was recorded in 17% of patients in the atazanavir/ritonavir group at week 96 compared with 23% at baseline, and in 27% of patients in the lopinavir/ritonavir group (also 27% at baseline). Fewer patients taking atazanavir/ritonavir (2%) than lopinavir/ritonavir (9%) initiated lipid-reduction therapy after start of study therapy.
The 96-week data from this noninferiority study confirm the similar virologic efficacy of once-daily ritonavir-boosted atazanavir to twice-daily ritonavir-boosted lopinavir, both in combination with tenofovir/emtricitabine, for the treatment of antiretroviral-naive HIV-1 adults that was previously established at week 48.9 The difference in response between the regimens in the ITT analysis in favor of the atazanavir/ritonavir-based regimen at 96 weeks was driven by a similar virologic response rate with a higher rate of discontinuations among patients taking lopinavir/ritonavir. This was based on a combination of adverse events and withdrawal of consent, both of which were higher among the patients in the lopinavir/ritonavir group compared with those in the atazanavir/ritonavir group. Both atazanavir/ritonavir and lopinavir/ritonavir, in combination with tenofovir/emtricitabine, had similarly high rates of adherence in this trial, as measured by the MACS survey (>80%) throughout this study, and thus adherence differences cannot explain the differences in response between the 2 groups.
The CASTLE study is the first comparative study to confirm the durability of the once-daily atazanavir/ritonavir regimen in treatment-naive patients through 96 weeks, with noninferiority to, and higher response rates than, twice-daily lopinavir/ritonavir.
Response rates in this 96-week study were consistent with those seen in other studies of atazanavir/ritonavir18 and lopinavir/ritonavir7,8,19,20 in treatment-naive patients, after taking differences in study designs and populations into consideration. Similarly, the efficacy results are consistent with those reported for other long-term studies of PIs.21-24 Both the atazanavir/ritonavir and lopinavir/ritonavir regimens resulted in similar increases in CD4 cell counts over 96 weeks. Response rates of HIV RNA <50 copies/mL were achieved at consistent levels across all baseline CD4 cell counts in both treatment groups, with results that were higher in the atazanavir/ritonavir group than the lopinavir/ritonavir group due to a higher rate of discontinuations among patients taking lopinavir/ritonavir. In addition, both regimens were effective in patients with high baseline HIV RNA; response rates were somewhat lower in patients with higher baseline HIV RNA in both treatment groups. More patients with advanced disease (CD4 <50 cells/μL and HIV RNA 100,000 copies/mL or more) responded among those treated with atazanavir/ritonavir than with lopinavir/ritonavir. A poorer response to lopinavir/ritonavir in patients with more advanced disease compared with those with lower HIV RNA and CD4 cell counts has also been reported in other studies.8,25,26
As expected, there was a low rate of virologic failure during treatment with these boosted PI regimens. These findings are consistent with data from other studies.7,8,20 The rates of development of resistance to PIs over the course of the study were low, with only a single patient in each treatment group with virologic failure at 96 weeks developing phenotypic resistance to a study PI. The emergence of NRTI substitutions was also low, with 5 patients in each treatment group developing phenotypic resistance to emtricitabine and 2 patients on lopinavir/ritonavir with phenotypic resistance to tenofovir disoproxil fumarate.
The main difference between the regimens in this study was related to their safety and tolerability. Adverse events and toxicity are the most common causes of treatment failure, and among the PIs gastrointestinal tolerability is an established risk for this outcome.27 In this study, more patients in the lopinavir/ritonavir group than the atazanavir/ritonavir group withdrew consent, and more discontinued treatment due to adverse events. Furthermore, the patients in the lopinavir/ritonavir group experienced a higher rate of gastrointestinal intolerability than those in the atazanavir/ritonavir group. Of note, no patients in the atazanavir/ritonavir arm discontinued treatment due to diarrhea, and a greater proportion of patients given lopinavir/ritonavir who remained in the study required antidiarrheal medication compared with those given atazanavir/ritonavir. Although patients treated with atazanavir/ritonavir had a higher rate of bilirubin abnormalities, only few (3) patients discontinued treatment as a result, none of them between weeks 48 and 96.
As expected, lopinavir/ritonavir resulted in higher increases in lipids compared with atazanavir/ritonavir. Drug-related dyslipidemia is associated with increased cardiovascular risk in patients on PI-containing regimens.6,28-30 Some PIs are known to increase total cholesterol and LDL cholesterol levels,31 with a dose-response effect from ritonavir, although as seen in this study, the lipid profile changes are not consistent among different PIs. Although the ritonavir daily dose in the lopinavir/ritonavir arm was twice as high as in the atazanavir/ritonavir arm, total exposure to ritonavir was comparable between the treatment arms [geometric mean ratio for area under the concentration curve over 24 hours (AUC0-24) was 0.84; 90% CI, 0.61-1.15].32 The similarity in AUC0-24 for ritonavir with both treatment regimens suggests that the higher rates of metabolic and gastrointestinal abnormalities observed in the twice-daily lopinavir/ritonavir arm compared with the atazanavir/ritonavir arm must be in part related to the lopinavir component of the fixed-dose lopinavir/ritonavir regimen.
No new or unexpected safety concerns emerged between 48 and 96 weeks, and the gastrointestinal and lipid safety profiles of atazanavir/ritonavir remained more favorable than those of lopinavir/ritonavir at 96 weeks. Events leading to study drug discontinuation in both regimens were consistent with the known side-effect profiles of the study drugs and comparison/backbone treatments.
The open-label design of this study is a limitation. In addition, patients in the lopinavir/ritonavir treatment group were only allowed to switch from the capsule to the tablet formulation after 48 weeks. Although this allowed for a direct comparison between atazanavir/ritonavir and a single formulation of lopinavir/ritonavir for the first 48 weeks of the study, the lack of comparative data against the newer tablet formulation of lopinavir/ritonavir throughout the study may be relevant, particularly as the capsule formulation is no longer available. However, several factors argue against the significance of any possible effects of the lopinavir/ritonavir formulation: first, only 3 patients discontinued treatment before week 48 because of a preference for the lopinavir/ritonavir tablet, and second, the 2 formulations have been shown to have similar safety and tolerability profiles.19 Finally, the similarly high adherence rates for both the atazanavir/ritonavir and lopinavir/ritonavir regimens before and after week 48 also suggest that pill burden and tolerability were not limiting factors in these treatment-naive patients. It is unlikely, therefore, that study outcomes were affected as a result of the limited availability of the lopinavir/ritonavir tablet formulation.
The results of this long-term comparative study reinforce the use of once-daily atazanavir/ritonavir as a recommended first-line treatment option for HIV-1 infection. Atazanavir/ritonavir is noninferior to ritonavir-boosted twice-daily lopinavir, but with a more favorable overall safety profile, and, importantly, a better lipid profile than lopinavir/ritonavir in treatment-naive patients. This regimen is therefore an appropriate therapeutic option for antiretroviral-naive HIV-1-infected patients, and is highly effective in patients with advanced disease, as measured by baseline low CD4 and high HIV RNA levels.