Original article Unboosted atazanavir-based therapy maintains control of HIV type-1 replication as effectively as a ritonavir-boosted regimen

Transcription

1 Antiviral Therapy : (doi: /IMP1666) Original article Unboosted atazanavir-based therapy maintains control of HIV type-1 replication as effectively as a ritonavir-boosted regimen Jade Ghosn 1, Giampiero Carosi 2, Santiago Moreno 3, Vadim Pokrovsky 4, Adriano Lazzarin 5, Gilles Pialoux 6, Jose Sanz-Moreno 7, Agnes Balogh 8, Eric Vandeloise 8, Sophie Biguenet 9, Ghislaine Leleu 9, Jean Francois Delfraissy 1 * 1 AP-HP, Department of Internal Medicine and Infectious Diseases, Bicêtre University Hospital, Le Kremlin-Bicêtre, France 2 Department of Infectious and Tropical Diseases, Spedali Civili, University of Brescia, Brescia, Italy 3 Department of Infectious Diseases, Hospital Ramón y Cajal, Universidad de Alcala, Madrid, Spain 4 Central Research Institute of Epidemiology, Ministry of Health, Moscow, Russia 5 Clinic of Infectious Diseases, Ospedale San Raffaele, Milan, Italy 6 Department of Infectious and Tropical Diseases, Hôpital Tenon, Paris, France 7 Hospital Universitario Príncipe De Asturias, Madrid, Spain 8 Bristol Myers Squibb, Braine L Alleud, Belgium 9 Bristol Myers Squibb, Paris, France *Corresponding author Jean-francois.Delfraissy@bct.aphp.fr Background: Triple combination therapy based on a ritonavir (RTV)-boosted protease inhibitor plus two nucleoside/nucleotide reverse transcriptase inhibitors (NRTIs) has improved outcomes in HIV type-1 (HIV-1)-infected patients. For patients unable to tolerate these regimens, alternative therapeutic approaches are needed. Methods: We report a comparative, open-label study in treatment-naive patients who underwent initial induction treatment with a triple combination including RTVboosted atazanavir (ATV; 300/100 mg once daily). Patients who achieved an HIV-1 viral load <50 copies/ml after the induction period were then randomized in the maintenance phase either to continue on current treatment or to switch to unboosted ATV 400 mg once daily (plus two NRTIs unchanged). Results: A total of 252 patients entered the induction phase, of whom 172 were eligible for randomization in the maintenance phase (ATV/RTV n=85 and ATV n=87). The unboosted ATV regimen demonstrated non-inferior efficacy to the ATV/RTV regimen with 78% and 75% of patients, respectively, maintaining virological suppression (HIV-1 RNA <50 copies/ml) up to week 48 after randomization (difference estimate 2.9, 95% confidence interval ). Time to virological failure and change from the end of the induction phase in mean CD4 + T-cell counts were also similar between the treatment arms. Although both regimens were well-tolerated, unboosted ATV was associated with fewer adverse events, fewer total bilirubin abnormalities and an improved lipid profile compared with ATV/RTV. Conclusions: An HIV-1 combined treatment regimen based on unboosted ATV is a feasible treatment option for patients with established virological control who are unable to tolerate triple combination therapy including ATV/RTV. Introduction Control of HIV type-1 (HIV-1) viral infection in the majority of patients can be achieved with the appropriate combination of antiretroviral therapies. Current international guidelines [1,2] for the treatment of adults recommend a combined regimen, usually constructed with three antiretroviral drugs, including two nucleoside/nucleotide reverse transcriptase inhibitors (NRTI) plus either one non-nucleoside reverse transcriptase inhibitor (NNRTI), or one protease inhibitor (PI) with low-dose ritonavir (RTV) for pharmacokinetic enhancement or boosting. Atazanavir (ATV) is an azapeptide PI that has been marketed in the US since 2003 for the treatment of HIV-1 infection in adults and which is also approved in 2010 International Medical Press (print) (online) 993

2 J Ghosn et al. Europe. The recommended dose of ATV in the US (with RTV boosting) is a dose of 300 mg of ATV with 100 mg of RTV once daily, or ATV 400 mg alone if the patient is unable to tolerate RTV [3,4]. There are, however, some patients for whom a triple combination regimen based on a boosted PI is associated with an excessive level of adverse events (AEs), leading to poor adherence and/ or treatment discontinuation. Switching to an NNRTIbased triple combination is the main alternative option in these circumstances; however, some patients receiving two NRTIs and an RTV-boosted PI cannot be switched to a NNRTI-based regimen for reasons such as the established presence of NNRTI resistance mutations, potential drug drug interactions and specific side effects. An alternative strategy would be to switch patients who are unable to tolerate RTV to an unboosted PI-based regimen. The Induction Maintenance with Atazanavir in HIV-naive Patients (InduMa) study was designed to first investigate whether switching to unboosted ATV was as effective as continuing on ATV/RTV in maintaining viral suppression in patients with fully suppressed HIV-1 replication after a week ATV/RTV-based induction regimen. In addition, we aimed to assess whether unboosted ATV was associated with a lower incidence of AEs than the boosted regimen and, therefore, might be a treatment option for patients unlikely to remain on a boosted PI-based antiretroviral regimen. Methods Study design The InduMa study (clinical trial registry number NCT ) was a multicentre, open-label, randomized, parallel group study consisting of two phases. The induction phase consisted of treatment of antiretroviral-naive, HIV-1-infected patients with ATV/RTV (300/100 mg once daily) plus two NRTIs for weeks. At the end of the induction phase, patients with a confirmed undetectable viral load (two consecutive titres of HIV-1 RNA<50 copies/ml recorded between week 16 and week 28) were eligible to enter the maintenance phase where they were randomized (1:1) either to continue to receive ATV/RTV (300/100 mg once daily) or to switch to unboosted ATV (400 mg once daily) for a further 48 weeks. All patients continued with their previous NRTI backbone throughout the study. Patients not eligible to enter the maintenance phase were offered the opportunity to continue induction treatment (ATV/RTV plus two NRTIs) through to the end of the study (rescue arm) and underwent the same assessments as patients in the maintenance phase. Patients Patients recruited into the study were HIV-1-infected treatment-naive men or women with a plasma HIV-1 RNA level of 5,000 copies/ml and a CD4 + T-cell count of 50 cells/mm 3 at screening. Key exclusion criteria were primary HIV infection, opportunistic infection or another medical condition requiring acute treatment at time of enrolment, grade 4 disturbance of glucose, electrolytes, transaminases or haematology at screening, or concomitant treatment with tenofovir. The use of proton pump inhibitors was prohibited in this study. The concomitant simultaneous administration of H2-receptor antagonists with unboosted ATV was prohibited; unboosted ATV had to be separated by at least 2 h before or 10 h after the dose of H2-receptor antagonist in the switch arm. Either simultaneous or temporally separated coadministration of ATV/RTV with H2-receptor antagonist was allowed in the continuation arm. Assessment procedures Physical examination, AE monitoring, screening of HIV-1 RNA titre (lower limit of detection 50 copies/ ml; Roche Amplicor 1.5 Ultrasensitive Assay; Roche Diagnostics, Basel, Switzerland), CD4 + T-cell count and blood chemistry were assessed at entry, at weeks 4, 16 and 24, at the end of the induction phase and at each scheduled visit (weeks 4, 12, 24, 36 and 48 [end]) during the maintenance phase. In addition, the fasting plasma lipid profile (total cholesterol, low-density lipoprotein [LDL] cholesterol, high-density lipoprotein [HDL] cholesterol, non-hdl cholesterol and triglycerides) was measured at baseline, at week 24 of the induction phase and at each scheduled visit during the maintenance phase. All HIV-1 RNA analyses were performed by a central laboratory (Quintiles Laboratories Europe, Livingston, UK). Both a central laboratory and site laboratories were used for other scheduled laboratory evaluations. For patients with confirmed HIV-1 RNA 400 copies/ml, the first sample exceeding 500 copies/ml (Monogram assay amplification limit) and the corresponding baseline sample, were sent for analysis of genotype substitutions and phenotypic sensitivity. Efficacy end points The primary end point of the study was a comparison of the proportion of patients in the two treatment groups (ATV/RTV and unboosted ATV) with a plasma HIV-1 RNA level <50 copies/ml sustained through week 48 of the maintenance phase (intent-to-treat [ITT] analysis; non-completion = failure). Secondary end points for the maintenance phase were the proportion of patients with HIV-1 RNA viral load <400 copies/ml through to week 48, time to treatment failure (confirmed viral load 50 copies/ml and 400 copies/ml on two consecutive determinations or study withdrawal for any reason) and change from baseline in CD4 + T-cell count. Additional secondary end points were changes from baseline in HIV-1 RNA International Medical Press

3 Unboosted atazanavir and HIV-1 control viral load and CD4 + T-cell count at the end of the induction phase. Outcomes for patients in the rescue arm were also recorded. Safety end points Safety end points were assessed by standard methods in both induction and maintenance phases and included the frequency and severity of AEs, discontinuations because of AEs and laboratory abnormalities. Changes from baseline in fasting lipids and the proportion of participants with National Cholesterol Education Program guided categories of fasting lipids over time were also recorded [5]. The frequency of serious adverse events (SAEs) and discontinuation because of AEs were also recorded for patients in the rescue arm. Statistical considerations Sample size and power The sample size of 178 randomized patients (89 on each regimen) was chosen to provide at least 80% power to demonstrate that a response rate (primary end point) of the unboosted ATV group is within a 15% margin of the response rate of the ATV/RTV group, assuming a two-sided 95% confidence interval (CI) for the difference in response rate between treatment regimens, a response rate of 85% on both treatment regimens and a margin of -15% for the difference in response rates between treatment regimens. Statistical methods Treatment failure was defined as confirmed viral load 50 copies/ml on two consecutive determinations, or study withdrawal for any reason. The unboosted ATV regimen was considered non-inferior to the ATV/RTV regimen if the lower limit of the 95% CI for the difference in proportions of patients maintaining HIV-1 RNA<50 copies/ml at week 48 of the maintenance phase was greater than -15%. Secondary end points were similarly assessed using a rebound threshold of 400 copies/ml. Time to treatment failure and to virological rebound was measured from the start of the maintenance phase and was estimated using Kaplan Meier curves and life tables. Treatment comparisons were based on a hazard ratio with a 95% CI derived from a Cox proportional hazards model with the treatment regimen as a covariate. Ethics The study was conducted in accordance with the Declaration of Helsinki and was consistent with the International Conference on Harmonization Good Clinical Practice and applicable European and US regulatory requirements. All study procedures were approved by local ethical committees, and all participating patients provided written informed consent. Results Patient disposition, demographic and baseline characteristics Patients were recruited from 28 sites in 9 countries in Europe (Estonia, France, Germany, Ireland, Italy, Latvia, Russia, Spain and the UK). Over 90% of the patients entering the maintenance phase were from sites in France, Italy, Russia and Spain. Of the 288 HIV-1- infected patients screened, 252 (88%) satisfied the entry criteria and entered the induction phase (Figure 1). During the induction phase, 12% discontinued the study; the most common reasons for discontinuing were AEs, being lost to follow-up and withdrawn consent. In the maintenance phase, discontinuation rates for the ATV/RTV and unboosted ATV groups before week 48 were 14% and 8%, respectively, with a further 1% and 2%, respectively, discontinuing at week 48. The most common reasons for withdrawal from this phase of the study were AEs (5% and 1%, respectively), pregnancy (2% in each group) and withdrawn consent (2% in each group). There were no clinically significant differences between randomized patients and the other 80 patients who did not enter the maintenance phase (AB and EV, data not shown), but non-randomized patients had a slightly higher median viral load and a slightly lower mean CD4 + T-cell count than randomized patients at baseline. Of the 50 non-randomized patients who chose to continue induction phase therapy (rescue arm), 41 (82%) continued to the week 48 visit, 36 (72%) eventually achieved viral suppression of HIV-1 RNA<50 copies/ml and 29 (58%) maintained this level of suppression through to week 48. Demographic and baseline data for patients who were randomized are shown in Table 1. The median baseline HIV-1 RNA level for patients entering the induction phase was 4.95 log 10 copies/ml, with 45% of the patients having baseline HIV-1 RNA 100,000 copies/ml. The treatment groups were well-balanced. The initial NRTI backbone combinations most frequently used during the induction phase were lamivudine plus abacavir (ABC) in 53% of patients in the ATV/RTV arm and 51% of patients in the ATV arm, and lamivudine plus zidovudine (AZT) in 31% of patients in the ATV/RTV arm and 33% of patients in the ATV arm. NRTI use during the induction phase was unchanged in 95% of patients. Efficacy Induction phase After starting induction phase therapy, viral load decreased rapidly, followed by a slower, sustained reduction; mean ±se reduction was 2.13 ±0.035 log 10 copies/ml at week 4 and 3.21 ±0.048 log 10 copies/ml at week 24. Recovery of the CD4 + T-cell count followed Antiviral Therapy

4 J Ghosn et al. Figure 1. Study design Screening (n=288) Enrolment in induction phase (n=252) ATV 300 mg once daily + RTV 100 mg once daily + 2 NRTIs for weeks At weeks 26 30: confirmed VL<50 copies/ml No (n=50) Discontinued (n=30): AE (n=9); lost to follow-up (n=5); consent withdrawn (n=4); no longer meets criteria (n=3); poor/non-compliance (n=2); death (n=1); lack of efficacy (n=1); pregnancy (n=1); other (n=4) Yes (n=172) Continue on ATV 300 mg once daily + RTV 100 mg once daily + 2 NRTIs Rescue arm (48 weeks) Completed (n=41) Randomization in maintenance phase Discontinued (n=9): AE (n=1); death (n=1); lack of efficacy (n=1); lost to follow-up (n=2); poor/non-compliance (n=3); consent withdrawn (n=1) Ratio 1:1 Switch regimen (n=87) Continuation regimen (n=85) 48 Weeks therapy, ATV 400 mg once daily, Background 2 NRTIs unchanged 48 Weeks therapy, ATV 300 mg once daily + RTV 100 mg once daily, Background 2 NRTIs unchanged Discontinued (n=9): Pregnancy (n=2); withdrawn consent (n=2); AE (n=1); lost to follow-up (n=1); poor/non-compliance (n=1); poor/non-compliance at week 48 visit (n=2) Completed (n=78) Completed (n=7) Discontinued (n=13): AE (n=4); pregnancy (n=2); consent withdrawn (n=2); lost to follow-up (n=1); poor/non-compliance (n=1); no longer meets criteria (n=1); poor/non-compliance at week 48 visit (n=1); other (n=1) AE, adverse event; ATV, atazanavir; NRTI, nucleoside reverse transcriptase inhibitor; RTV, ritonavir; VL, viral load International Medical Press

5 Unboosted atazanavir and HIV-1 control Table 1. Demographics and baseline HIV-1 status of the randomized population Characteristic ATV/RTV (n=85) ATV (n=87) Total (n=172) Median age, years (range) 35 (19 70) 35 (21 66) 35 (19 70) Male, n (%) 61 (72) 65 (75) 126 (73) Race White, n (%) 70 (82) 73 (84) 143 (83) Black/mixed, n (%) 14 (16) 14 (16) 28 (16) Other, n (%) 1 (1) 0 1 (0.6) Median BMI at end of induction, kg/m 2 (range) 23.1 ( ) 23.1 ( ) 23.1 ( ) Prior CDC Class C AIDS events, n (%) 2 (2) 0 2 (1) Hepatitis B or C coinfection, n (%) 20 (24) 16 (19) 36 (22) Median HIV-1 RNA viral load at baseline, log 10 copies/ml (range) 4.86 ( ) 4.85 ( ) 4.85 ( ) Median CD4 + T-cell count at baseline, cells/mm 3 (range) 265 (64 490) 255 (50 660) 259 (50 660) Median time on induction phase therapy was 28 weeks. Nucleoside reverse transcriptase inhibitor (NRTI) use throughout the induction phase was unchanged in 95% of patients. The most frequently used NRTI backbone combinations by patients were lamivudine plus abacavir (50%) and lamivudine plus zidovudine (32%). ATV, atazanavir; ATV/RTV, ritonavir-boosted ATV; BMI, body mass index; HIV-1, HIV type-1. Figure 2. Proportion of patients with HIV-1 control for the ITT population and observed cases ATV/RTV (n=85) ATV (n=87) A Patients with VL<50 copies/ml, % ( ) a 75% 78% ITT (NC=F) Primary end point -1.5 ( ) a 89% 88% Observed b Patients with VL<400 copies/ml, % B ( ) a 81% 86% ITT (NC=F) Primary end point -0.8 ( ) a 95% 94% Observed b Randomized patients with (A) HIV type-1 (HIV-1) RNA<50 copies/ml and (B) HIV-1 RNA<400 copies/ml are shown. a Difference estimate (95% confidence intervals). b Observed cases are based on single, rather than sequential HIV-1 RNA values and expressed as a proportion of patients remaining on active treatment at the time of the scheduled visit. ATV, atazanavir; ATV/RTV, ritonavir-boosted ATV; ITT, intent-to-treat; NC=F, non-completion equals failure; VL, viral load. a similar time course, increasing by a mean ±se of 88 ±6.1 cells/mm 3 by week 4 and 177 ±8.8 cells/mm 3 by week 24. Randomized maintenance phase ATV had non-inferior antiviral efficacy compared with ATV/RTV through to week 48 of the maintenance phase. There was no significant difference in the proportion of patients with HIV-1 RNA remaining <50 copies/ml in either the ITT population (primary analysis, 75% of patients on ATV/RTV compared with 78% on ATV [difference estimate 2.9%, 95% CI ; Figure 2]), or in observed cases (based on single rather than sequential HIV-1 RNA values and expressed as a proportion of patients remaining on active treatment at the time of the scheduled visit) where 89% of patients were on ATV/RTV compared with 88% on ATV (difference estimate -1.5%, 95% CI ; Figure 2). The proportions of patients with HIV-1 RNA<50 copies/ml were also consistent within and between treatment arms for all scheduled visits throughout the maintenance phase for both the ITT and observed case Antiviral Therapy

6 J Ghosn et al. populations. Corresponding values for patients based on maintaining a viral load of <400 copies/ml supported the non- inferiority of unboosted ATV versus ATV/RTV (Figure 2). Discontinuation for any reason during the maintenance phase was classified as lack of treatment success and occurred in 14% of patients receiving ATV/RTV and 8% receiving unboosted ATV. Virological rebound rates up to week 48, defined as HIV-1 RNA 50 copies/ml, were 7% for the ATV/RTV regimen and 13% for the unboosted ATV regimen. Corresponding rates for patients with viral loads 400 copies/ml were 1% and 5%, respectively. The time to treatment failure was consistent between the ATV/ RTV and ATV regimens for both HIV-1 RNA 50 copies/ml and HIV-1 RNA 400 copies/ml; hazard ratios (ATV:ATV/RTV) and 95% CIs for time to treatment failure were 0.97 ( ) for HIV-1 RNA 50 copies/ml and 0.84 ( ) for HIV-1 RNA 400 copies/ml. The time to virological rebound was comparable between the two regimens (hazard ratios [95% CIs] for virological rebound were 1.41 [ ] for HIV-1 RNA 50 copies/ml and 3.55 [ ] for HIV-1 RNA 400 copies/ml), although these data do not exclude a difference between the two regimens in terms of virological rebound. Regimens were also comparable for mean change in CD4 + T-cell count between the end of the induction phase at week 48 and the maintenance phase; the mean ±se change was 92 ±18 cells in the ATV/RTV arm and 100 ±15 cells in the ATV arm. Development of drug resistance Of the 17 patients (6 on the ATV/RTV regimen and 11 on the unboosted ATV regimen) with virological rebound >50 copies/ml, 2 and 4 patients, respectively, had a viral load >500 copies/ml during the maintenance phase. Analysis of these latter samples and corresponding baseline samples showed no phenotypic resistance to ATV or other PIs and genotypic resistance testing detected an M184V substitution in reverse transcriptase in three patients (one in the ATV/RTV group and two in the ATV group). Safety Deaths and other serious adverse events There were no unexpected safety events. A total of four deaths were reported, all on ATV/RTV treatment, three during the induction phase and one in the rescue arm. One death, caused by pneumonia, pancytopaenia, gastroenteritis, pancreatitis and multiorgan failure, was considered possibly related to study medication. The other three deaths were considered unrelated or unlikely to be related to treatment. There were no deaths during the maintenance phase. Among patients receiving ATV/RTV, SAEs were experienced by 16 (6%) patients during the induction phase, 3 (4%) in the maintenance phase and 2 (4%) in the rescue arm (Table 2). A total of four (5%) patients receiving unboosted ATV reported SAEs. The events were consistent with the known side effect profiles of the study drugs. In total, six SAEs (five possible and one probable) in two patients were considered related to study medication. No SAEs of jaundice were reported during the study. Adverse events leading to discontinuation of study therapy There were nine discontinuations because of AEs in patients during the induction phase and one in the rescue arm. During the maintenance phase, four patients receiving ATV/RTV and one receiving unboosted ATV discontinued study therapy. Two (2%) patients in the ATV/RTV group discontinued because of hyperbilirubinaemia or jaundice; however, there was no consistency in the nature of AEs leading to withdrawal from the study. Overall adverse events AEs are shown in Table 2; the majority were mild-tomoderate (grade 1 to grade 2). AEs that differed in frequency between the treatment groups (ATV/RTV versus unboosted ATV) by 5% included increases in bilirubin (26% versus 15%) and aspartate aminotransferase (1% versus 6%); however, routine laboratory testing showed that for both alanine aminotransferase and aspartate aminotransferase, the incidence of grades 1 4, grades 3 4 and grade 4 abnormalities were consistent between regimens. Changes in fasting lipids and fasting glucose After randomization, there was a clinically significant decrease in all atherogenic lipids from the end of the induction phase to week 48 of the maintenance phase in the unboosted ATV group (Table 3). In addition, there was a statistically significant difference in favour of the unboosted ATV group in the change in fasting triglycerides: 9.8% and -27.0% for the ATV/RTV and unboosted ATV groups, respectively (P<0.0001). More patients receiving ATV/RTV than those receiving unboosted ATV moved into higher National Cholesterol Education Program categories for total cholesterol (23% and 10%, respectively) and triglycerides (20% and 3%, respectively). Fasting glucose abnormalities were reported in 11% and 9% of patients for ATV/RTV and unboosted ATV regimens, respectively. Grade 3 4 abnormalities occurred in 1% of patients in each group. Discussion The current study was an open-label randomized design consisting of two phases and was designed to examine International Medical Press

7 Unboosted atazanavir and HIV-1 control Table 2. Adverse events of clinical interest, of all grades, or with a frequency of 5% reported in induction and maintenance phases Induction phase Maintenance phase AE ATV/RTV (n=252) ATV/RTV (n=85) ATV (n=87) Deaths, n (%) 3 (1) 0 0 Total SAEs, n (%) 16 (6) 3 (4) 4 (5) SAEs related to study therapy, n (%) 2 (0.8) 0 0 AEs leading to discontinuation, n (%) 9 (4) 4 (5) 1 (1) Total AEs, all grades, n (%) 203 (81) 74 (87) 67 (77) Bilirubin-related disorders, n (%) 107 (42) 28 (33) 18 (21) Hyperbilirubinaemia, n (%) a 81 (32) 24 (28) 14 (16) Blood bilirubin increased, n (%) 76 (30) 22 (26) 13 (15) Jaundice, n (%) 14 (6) 3 (4) 0 Ocular icterus, n (%) 32 (13) 6 (7) 4 (5) Infections and infestations, n (%) 64 (25) 30 (35) 29 (33) Bronchitis, n (%) 8 (3) 3 (4) 6 (7) Nasopharyngitis, n (%) 7 (3) 5 (6) 4 (5) Laboratory investigations, n (%) 87 (35) 36 (42) 23 (26) ALT increased, n (%) 17 (7) 5 (6) 4 (5) AST increased, n (%) 13 (5) 1 (1) 5 (6) Gastrointestinal disorders, n (%) 78 (31) 21 (25) 19 (22) Nausea, n (%) 27 (11) 4 (5) 4 (5) Diarrhoea, n (%) 22 (9) 7 (8) 5 (6) Abdominal pain, n (%) 8 (3) 3 (4) 5 (6) General/administrative site conditions, n (%) 40 (16) 11 (13) 6 (7) Asthaenia, n (%) 20 (8) 4 (5) 2 (2) Nervous system disorders, n (%) 26 (10) 14 (16) 9 (10) Headache, n (%) 16 (6) 8 (9) 7 (8) Skin and subcutaneous disorders, n (%) 34 (13) 7 (8) 7 (8) The rescue arm contained one death, two serious adverse events (SAEs) and one discontinuation because of adverse events (AEs). a Includes the preferred terms blood bilirubin increased, blood bilirubin abnormal, blood bilirubin unconjugated and blood bilirubin unconjugated increased. ALT, alanine aminotransferase; AST, aspartate aminotransferase; ATV, atazanavir; ATV/RTV, ritonavir-boosted ATV. Table 3. Fasting lipid mean percentage changes from end of induction at week 48 a in patients treated with maintenance phase therapy ATV/RTV b ATV c Mean value Mean % change Mean value Mean % change Lipids at EoI (±se) at week 48 (95% CI) at EoI (±se) at week 48 (95% CI) P-value d Total cholesterol 188 (4.3) 1.4 ( ) 192 (5.1) -4.7 ( ) HDL cholesterol 51 (1.6) 0.8 ( ) 53 (1.8) 3.7 ( ) 0.42 LDL cholesterol 109 (3.7) -2.1 ( ) 109 (4.3) -0.7 ( ) 0.74 Triglycerides 138 (8.9) 9.8 ( ) 164 (18.0) ( ) < Non-HDL cholesterol 137 (4.4) 1.1 ( ) 140 (5.1) -7.4 ( ) Values are excluded after the start of serum lipid reduction therapy. Because of multiple comparisons, only P-values <0.005 were considered statistically significant. a Last observation carried forward. b Number of ritonavir-boosted atazanavir (ATV/RTV) participants is 77 for all parameters. c Number of atazanavir (ATV) participants is 81 for all parameters. d P-values for the difference between percentage changes in ATV/RTV compared with ATV are shown. CI, confidence interval; EoI, end of induction phase; HDL, high-density lipoprotein; LDL, low-density lipoprotein. the efficacy, safety and tolerability of an unboosted ATV-based regimen compared with an ATV/RTV regimen, both after initial induction with boosted ATV (NRTIs unchanged throughout the study). The primary efficacy analysis demonstrated that over the course of the study, the unboosted ATV regimen had non-inferior antiviral efficacy to the boosted regimen, a standard-of-care treatment combination recommended by current treatment guidelines [1,2]. Similar to several other studies in antiretroviral-naive patients conducted during the same time period as InduMa [6,7], the study design allowed proven non-inferiority if the value for the lower limit of the 95% CI was >-15%, rather than the 10 12% difference that has been used in several other Antiviral Therapy

8 J Ghosn et al. studies. The actual value of the lower limit in this study was -9.8%. Other (secondary) efficacy results, including the proportion of patients with HIV-1 RNA<400 copies/ml and recovery of CD4 + T-cell count from baseline, were consistent with and further supported the primary analysis. No evidence of emerging PI resistance was detected in the six patients (four ATV and two ATV/ RTV) with viral load >500 copies/ml during the maintenance phase, suggesting that future PI treatment options for these patients were not compromised by treatment with unboosted ATV. Another objective of the study was to determine whether unboosted ATV was better tolerated than the boosted regimen. Overall discontinuation rates were lower on the unboosted ATV regimen than on the ATV/ RTV regimen; discontinuation rates from the latter regimen were consistent with those previously observed for ATV/RTV in treatment-naive patients [8]. The pattern of AEs was in keeping with the known safety profile of ATV, RTV and the backbone NRTIs reported in previous studies with ATV [8 11]. The overall discontinuation rates in the current study suggest that ATV is better tolerated than ATV/RTV. Boosted ATV is known to be associated with an improved plasma lipid profile compared with other boosted PIs [8,11 15]. Interestingly, in this study, patients who switched to unboosted ATV experienced significantly greater improvements in triglyceride levels than those continuing the boosted regimen. In addition, the unboosted ATV group had a lower incidence of grades 1 4, grades 3 4 and grade 4 total bilirubin abnormalities than the ATV/RTV group. These findings are consistent with a study that compared boosted and unboosted ATV in treatment-naive patients [16]. Furthermore, grades 3 4 total bilirubin abnormalities experienced by a small number of patients at the end of the induction phase improved during the maintenance phase in more patients receiving unboosted ATV than ATV/RTV. The use of NRTIs was balanced between the ATV and ATV/RTV treatment arms, with the majority of patients overall receiving a combination including ABC, and nearly one-third of patients overall receiving a combination including AZT. Recent studies have reported an association between ABC therapy and cardiovascular risk in HIV-infected patients [17,18] and between AZT therapy and lipoatrophy [19]; such potential associations should be considered during treatment decision making. Since the completion of the InduMa trial, other treatment options have become available, including raltegravir, an HIV integrase inhibitor [20], which might allow for an NRTI-sparing regimen [21] and has demonstrated efficacy compared with background treatment alone in patients with drug-resistant HIV infection [22]. Study limitations The current study has several potential limitations. Firstly, it might be argued that the median 28-week duration of the induction phase was too short because 58% of patients in the rescue arm eventually achieved and maintained HIV-1 RNA<50 copies/ml. At the time of the development of the study protocol, the length of the induction phase used in comparable studies was commonly a minimum of 24 weeks, with a median of 24 weeks induction in a study comparing ritonavir-boosted lopinavir with efavirenz [23], and up to 32 weeks in the Forte trial [24]. In the InduMa study, the induction phase was weeks. An analysis of time to suppression (AB and EV, data not shown) suggests that an induction phase of approximately 40 weeks would have been required to capture all of the potential responders. Although there were too few patients for a definitive analysis, the time to achieve a confirmed HIV-1 RNA titre of <50 copies/ml might have been related to baseline values. It might, therefore, be argued that there was a bias towards selecting patients with lower baseline titres at entry for the randomized maintenance phase, although data from another study of similar design but with a longer induction phase (36 weeks) suggested that non-inferiority of unboosted ATV relative to ATV/RTV was not influenced by viral load at baseline [25]. Secondly, triple combination therapy is a long-term treatment. Results at week 48 of the maintenance phase of the study might not be predictive of the long-term efficacy of the unboosted ATV regimen. Similarly, the timescale of the study does not permit detection of any differences between regimens in the rate of emergent resistant phenotypes, although the observed frequency of virological rebound was slightly higher in patients receiving the unboosted regimen. Finally, recent data suggest that among NRTIs, a tenofovir-based triple combination is associated with good clinical efficacy and tolerability [26], particularly in patients with HIV RNA>100,000 copies/ml before treatment [27] and a low incidence of lipid disturbance [28]. Consequently, this regimen might be used more extensively in accordance with current international guidelines [1,2]. Although boosted ATV and tenofovir is an effective treatment combination [29], drug interactions preclude the use of unboosted ATV with tenofovir [30]. This might present a further limitation on translating the results of the InduMa study into clinical practice because it would also be necessary to change the NRTI backbone when switching tenofovir-treated patients, with an unacceptable level of side effects, from boosted to unboosted ATV. Raltegravir might present a suitable alternative to tenofovir in this context, but this is beyond the scope of this study. In conclusion, previous studies with ATV but without the coadministration of RTV have suggested that International Medical Press

9 Unboosted atazanavir and HIV-1 control a regimen of 400 mg ATV (plus an NRTI backbone) was safe and not inferior to nelfinavir and efavirenz in treatment-naive patients for up to 72 weeks, and also resulted in minimal changes in lipid metabolism. More recently, ATV demonstrated non-inferior efficacy (timeto-loss of virological response, that is <50 copies/ml) to ATV/RTV at the same doses used in the current study at week 48 after randomization [15]. Both treatments were administered in combination with ABC/lamivudine, and the treatment effects were independent of baseline viral load. The current study demonstrates that after induction with boosted ATV, switching to unboosted ATV shows similar (non-inferior) efficacy and a more favourable safety profile than a triple combination regimen based on boosted ATV for up to 48 weeks. However, longerterm studies are needed to demonstrate that unboosted ATV can provide sustained efficacy comparable with that of the boosted regimen; therefore treatment-naive patients should continue to receive a first-line ATV/ RTV-based triple combination regimen, as advocated by treatment guidelines. For patients unable to tolerate ritonavir, switching to unboosted ATV might represent a feasible alternative treatment option. Acknowledgements The authors would like to thank the investigators who took part in this trial. A full list of these investigators can be found via Additional file 1. Financial support was provided by Bristol Myers Squibb. Assistance with the study design, monitoring and drug supplies for the study were provided by Bristol Myers Squibb. Data collation and statistical analyses were performed by Bristol Myers Squibb. The investigators had full access to the data and the opinions expressed in this paper are those of the investigators. The authors would like to thank Les Steele of Prism Ideas (Nantwich, UK) for editorial support in the generation of this manuscript. Disclosure statement JG is part of the board of Bristol Myers Squibb, Gilead Sciences, MSD and Tibotec (a division of Janssen Cilag). SM has received grant support or has participated with presentations in activities organized by Abbott Laboratories, Bristol Myers Squibb, Glaxo- SmithKline, Roche, MSD, Janssen Cilag, Pfizer, Gilead Sciences, Boehringer Ingelheim and Schering Plough. JS-M has received financial support from Bristol Myers Squibb, Abbott Laboratories, Gilead Sciences, Boehringer Ingelheim and GlaxoSmithKline in the form of either research grants or speaker s honoraria. GC has received grants for attending expert meetings, has received speaker s honoraria and grants for research from various pharmaceutical companies including Bristol Myers Squibb, Roche, Pfizer and Gilead Sciences. AL has received consulting and lecture fees from Abbott Laboratories, GlaxoSmithKline, Bristol Myers Squibb, Janssen Cilag, Roche, Boehringer Ingelheim, Gilead Sciences and Schering Plough. GP is part of the board of Bristol Myers Squibb, Tibotec (a division of Janssen Cilag), Roche and Schering Plough. J-FD is a member of the International Advisory Boards of Roche, Bristol Myers Squibb, Gilead Sciences and Tibotec, and has collaborated on clinical studies with Abbott Laboratories and GlaxoSmithKline. AB, EV, GL and SB are employees of Bristol Myers Squibb. VP declares no competing interests. Additional file Additional file 1: A list of investigators who took part in this trial can be found at com/uploads/documents/avt-10-oa-1528_ghosn_ Add_file1.pdf References 1. Panel on Antiretroviral Guidelines for Adults and Adolescents. Guidelines for the use of antiretroviral agents in HIV-1-infected adults and adolescents. US Department of Health and Human Services. (Updated 3 November Accessed September 2009.) Available from aidsinfo.nih.gov/contentfiles/adultandadolescentgl.pdf 2. Hammer SM, Eron JJ, Jr., Reiss P, et al. Antiretroviral treatment of adult HIV infection: 2008 recommendations of the International AIDS Society USA panel. JAMA 2008; 300: Reyataz (atazanavir sulfate). Package Insert Bristol Myers Squibb, Uxbridge, UK. 4. Reyataz (atazanavir sulfate). Highlights of prescribing information. Package Insert Bristol Myers Squibb, Uxbridge, UK. (Updated September Accessed 08 September 2010.) Available from fda.gov/drugsatfda_docs/label/2008/021567s017lbl.pdf 5. Third report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III) final report. Circulation 2002; 106: Johnson MA, Gathe JC, Jr., Podzamczer D, et al. A oncedaily lopinavir/ritonavir-based regimen provides noninferior antiviral activity compared with a twice-daily regimen. J Acquir Immune Defic Syndr 2006; 43: Cooper D, Zajdenverg R, Ruxrungtham K, Chavez RL. Efficacy and safety of two doses of tipranavir/ritonavir versus lopinavir/ritonavir-based therapy in antiretroviralnaive patients: results of BI HIV8, 8th International Congress on Drug Therapy in HIV infection November 2006, Glasgow, UK. Abstract Molina JM, Andrade-Villanueva J, Echevarria J, et al. Once-daily atazanavir/ritonavir versus twice-daily lopinavir/ritonavir, each in combination with tenofovir and emtricitabine, for management of antiretroviral-naive HIV- 1-infected patients: 48 week efficacy and safety results of the CASTLE study. Lancet 2008; 372: Johnson M, Grinsztejn B, Rodriguez C, et al. 96-Week comparison of once-daily atazanavir/ritonavir and twicedaily lopinavir/ritonavir in patients with multiple virologic failures. AIDS 2006; 20: Antiviral Therapy

10 J Ghosn et al. 10. Soriano V, Garcia-Gasco P, Vispo E, et al. Efficacy and safety of replacing lopinavir with atazanavir in HIVinfected patients with undetectable plasma viraemia: final results of the SLOAT trial. J Antimicrob Chemother 2008; 61: Mallolas J, Podzamczer D, Milinkovic A, et al. Efficacy and safety of switching from boosted lopinavir to boosted atazanavir in patients with virological suppression receiving a LPV/r-containing HAART: the ATAZIP study. J Acquir Immune Defic Syndr 2009; 51: Gatell J, Salmon-Ceron D, Lazzarin A, et al. Efficacy and safety of atazanavir-based highly active antiretroviral therapy in patients with virologic suppression switched from a stable, boosted or unboosted protease inhibitor treatment regimen: the SWAN Study (AI ) 48-week results. Clin Infect Dis 2007; 44: Mobius U, Lubach-Ruitman M, Castro-Frenzel B, et al. Switching to atazanavir improves metabolic disorders in antiretroviral-experienced patients with severe hyperlipidemia. J Acquir Immune Defic Syndr 2005; 39: Jackson A, Patel N, Lo G, Gazzard B, Moyle G. Effects of atazanavir or saquinavir once daily with ritonavir 100mg and tenofovir/emtricitabine as initial therapy for HIV-1 infection on peripheral glucose disposal: a randomized open-label study. 14th Conference on Retroviruses and Opportunistic Infections February 2007, Los Angeles, CA, USA. Abstract Stanley TL, Joy T, Hadigan CM, et al. Effects of switching from lopinavir/ritonavir to atazanavir/ritonavir on muscle glucose uptake and visceral fat in HIV-infected patients. AIDS 2009; 23: Malan DR, Krantz E, David N, Wirtz V, Hammond J, McGrath D. Efficacy and safety of atazanavir, with or without ritonavir, as part of once-daily highly active antiretroviral therapy regimens in antiretroviral-naive patients. J Acquir Immune Defic Syndr 2008; 47: Behrens GM, Reiss P. Abacavir and caridiovascular risk. Curr Opin Infect Dis 2010; 23: Strategies for Management of Anti-Retroviral Therapy/ INSIGHT and the DAD Study Groups. Use of nucleoside reverse transcriptase inhibitors and risk of myocardial infarction in HIV-infected patients. AIDS 2008; 22:F17 F van Vonderen MG, van Agtmael MA, Hassink EA, et al. Zidovudine/lamivudine for HIV-1 infection contributes to limb fat loss. PLoS ONE 2009; 4:e Garrido C, Soriano V, de Mendoza C. New therapeutic strategies for raltegravir. J Antimicrob Chemother 2010; 65: Allavena C, Mounoury O, Rodallec A, et al. Efficacy and safety of an NRTI-sparing dual regimen of raltegravir and ritonavir-boosted protease inhibitor in a triple antiretroviral class-experienced population. HIV Clin Trials 2009; 10: Steigbigel RT, Cooper DA, Teppler H, et al. Long-term efficacy and safety of raltegravir combined with optimized background therapy in treatment-experienced patients with drug-resistant HIV infection: week 96 results of the BENCHMRK 1 and 2 Phase III trials. Clin Infect Dis 2010; 50: Cameron DW, da Silva BA, Arribas JR, et al. A 96-week comparison of lopinavir-ritonavir combination therapy followed by lopinavir-ritonavir monotherapy versus efavirenz combination therapy. J Infect Dis 2008; 198: Asboe D, Williams IG, Goodall RL, Darbyshire JH, Hooker MH, Babiker AG. A virological benefit from an induction/maintenance strategy: the Forte trial. Antivir Ther 2007; 12: Squires K. Similar efficacy and tolerability of atazanavir (ATV) compared to ATV/ritonavir (RTV, r), each in combination with abacavir/lamivudine (ABC/3TC), after initial suppression with ABC/3TC + ATV/r in HIV-1 infected patients: 84 week results of the ARIES trial. 5th IAS Conference on HIV Pathogenesis, Treatment and Prevention July 2009, Cape Town, South Africa. Abstract WELBB Gallant JE, DeJesus E, Arribas JR, et al. Tenofovir DF, emtricitabine, and efavirenz vs. zidovudine, lamivudine, and efavirenz for HIV. N Engl J Med 2006; 354: Sax P, Tierney C, Collier A, et al. ACTG 5202: shorter time to virologic failure (VF) with abacavir/lamivudine (ABC/3TC) than tenofovir/emtricitabine (TDF/FTC) as part of combination therapy in treatment-naive subjects with screening HIV RNA 100,000 c/ml. 17th International AIDS Conference. 3 8 August 2008, Mexico City, Mexico. Abstract THAB Valantin MA, de Truchis P, Bittar R, et al. Switching NRTI backbone to tenofovir plus emtricitabine (Truvada) promptly improves triglycerides and LDL-cholesterol levels in dyslipidemic HIV patients: the TOTEM randomised trial. 17th International AIDS Conference. 3 8 August 2008, Mexico City, Mexico. Abstract THPE Soriano V, Köppe S, Mingrone H, et al. Prospective comparison of nevirapine and atazanavir/ritonavir both combined with tenofovir DF/emtricitabine in treatmentnaive HIV-1 infected patients: ARTEN study week 48 results. 5th IAS Conference on HIV Pathogenesis, Treatment and Prevention July 2009, Cape Town, South Africa. Abstract LBPE Electronic Medicines Compendium. Reyataz 150 mg, 200 mg and 300 mg hard capsules. (Updated 12 August Accessed 8 September 2010.) Available from Accepted for publication 17 April International Medical Press