Clinical Trial Result Information

Protocol number
ML18413

Title of Study
A 48-week, randomized, open-label, 2-arm study to compare the efficacy of saquinavir/ritonavir BID plus emtricitabine/tenofovir QD versus lopinavir/ritonavir BID plus emtricitabine/tenofovir QD in treatment-naive HIV-1 infected patients (Gemini Study)

Sponsor
Roche Laboratories Inc

Company division
Pharmaceutical

Product name
Invirase

Generic name
saquinavir

Therapeutic area
HIV infections

Clinical study summary
This was a prospective, multicenter, randomized, open-label, 2-arm study to compare the efficacy and safety of saquinavir/ritonavir BID plus emtricitabine/tenofovir QD versus lopinavir/ritonavir BID plus emtricitabine/tenofovir QD in treatment-naïve HIV-1 infected patients. It was a 48-week study, with a planned analysis at 24 weeks.

Study center(s)
38 centers in Canada, France, Thailand and the United States.

Phase of development
IV

Objectives
Primary: To evaluate the efficacy of saquinavir/ritonavir BID plus emtricitabine/tenofovir QD versus lopinavir/ritonavir BID plus emtricitabine/tenofovir QD in treatment-naive HIV-1 infected adults.
Secondary: To evaluate the safety, adherence, and tolerability of saquinavir/ritonavir BID plus emtricitabine/tenofovir QD versus lopinavir/ritonavir BID plus emtricitabine/tenofovir QD in treatment-naive HIV-1 infected adults.

Methodology
At the baseline visit, eligible patients were randomized in a 1:1 ratio to one of two treatment arms: saquinavir/ritonavir 1000/100mg  BID plus emtricitabine/tenofovir 200/300mg QD (arm 1) or lopinavir/ritonavir 400/100mg BID plus emtricitabine/tenofovir 200/300mg QD (arm 2).

Efficacy, tolerability and safety assessments took place at clinic visits at intervals up to the end of week 48. Patients returned for a follow-up safety visit 4 weeks after the end of treatment.

Number of patients (planned/analyzed)
Planned: 350. Analyzed: 337-ITT, 331-safety.

Diagnosis and main criteria for inclusion
Chronically HIV-1 infected, treatment-naive adults (2 weeks prior ARV therapy); HIV-1 RNA >10,000 copies/mL, CD4+ count ≤350 cells/mL.

Test product, dose and mode of administration or test procedure
Saquinavir/ritonavir 1000/100mg po BID plus emtricitabine/tenofovir 200/300mg po QD.

Duration of treatment
48 weeks

Reference therapy, dose and mode of administration or reference procedure
Lopinavir/ritonavir 400/100mg po BID plus emtricitabine/tenofovir 200/300mg po QD.

Criteria for evaluation (efficacy, safety)
Efficacy: Primary: Number and percentage of patients with an HIV-1 RNA viral load <50 copies/mL at week 48.
Secondary: At weeks 2, 4, 8, 12, 16, 20, 24, 36, and 48, number and percentage of patients with an HIV-1 RNA viral load <50 copies/mL, change from baseline in log transformed HIV-1 RNA viral load (copies/mL), number and percentage of patients with an HIV-1 RNA viral load <400 copies/mL and change from baseline in CD4+ lymphocyte count; virologic failure, defined as any 2 consecutive measurements (≥14 days apart) of HIV-1 RNA >400 copies/mL at week 16 or thereafter; time to loss of virologic response (TLOVR); number and percentage of patients with hypertension at weeks 4, 8, 12, 16, 20, 24, 36 and 48; number and percentage of patients discontinuing study medication due to clinical adverse events (including clinically significant laboratory abnormalities and ACTG grade ≥2 laboratory toxicities); number and percentage of patients adhering to antiretroviral treatment regimen at weeks 2, 4, 8, 12, 16, 20, 24, 36, and 48.

Safety: Adverse events monitored throughout the study. AIDS-defining events (weeks 24 and 48). Physical examinations, vital signs, and clinical laboratory tests. 
 

Statistical methods
The final analysis was based on 1-sided non-inferiority hypothesis testing at alpha = 0.020:
In the final analysis, with a sample size of 310, there was 81% power to claim non-inferiority in the saquinavir arm compared to the lopinavir arm, if the actual difference was 0% (85% vs. 85%).

Efficacy Analyses
The proportion of patients with viral load <50 copies/mL was assessed for non-inferiority constructing 1-sided 98% confidence intervals (CIs) using the normal approximation to the binomial distribution. Non-inferiority of arm 1 to arm 2 was established if the lower limit of the difference (arm 1 - arm 2) in the response rate was not equal to or less than - 12%.
Statistical tests of the null hypothesis of no treatment difference between arm 1 and arm 2 was also conducted for the proportion of patients with viral load <50 copies/mL and <400 copies/mL by study week, the proportion of patients with change from baseline (ie, decrease) in log₁₀ HIV-1 RNA viral load < -0.5 log₁₀ and < -1.0 log₁₀ at week 48, proportion of patients with viral failure, and the proportion of patients with hypertension.
An analysis of covariance model (ANCOVA) was used to model the treatment difference in the change from baseline in log₁₀ plasma HIV-1 RNA viral load at weeks 24 and 48. Treatment arms were compared for change from baseline in CD4+ lymphocyte count using the Wilcoxon Rank Sum test. The Kaplan-Meier estimate was used to estimate the time to virologic response (first HIV RNA <400 copies/mL) and time to loss of virologic response.

Summary (efficacy, safety, other results)
Efficacy: In the primary efficacy analysis, 64.7% of patients in the saquinavir/r arm and 63.5% of patients in the lopinavir/r arm had an HIV-1 RNA viral load <50 copies/mL at week 48 for a difference of 1.14% (96% CI = -9.6%, 11.9%). Since the lower limit of the 2-sided 96% CI for the between-group difference in response rate was not equal to or less than -12%, non-inferiority of saquinavir/r to lopinavir/r was established.
Week 48 virologic and immunologic response data (ITT populations) are as follows:

	Saquinavir/r	Lopinavir/r
No. (%) patients with an HIV-1 RNA <50 copies/mL1,2	64.7%	63.5%
No. (%) patients with an HIV-1 RNA <400 copies/mL1,2	72.5%	74.7%
Mean change from baseline in log10 HIV-1 RNA (copies/mL)2	-3.39	-3.36
Median change from baseline in CD4+ count (/mm3)2	178.0	204.0

1. Missing values imputed as nonresponse
2. Difference between treatment arms not statistically significant

Eleven (6.6%) patients in the saquinavir/r arm and 5 (2.9%) patients in the lopinavir/r arm had confirmed virologic failure, defined as any 2 consecutive measurements (≥14 days apart) of HIV-1 RNA >400 copies/mL at or after week 16 (P=0.1311). Seven patients who experienced virologic failure, 6 in saquinavir/r arm and 1 in the lopinavir/r arm, had no new HIV mutations at virologic failure compared to baseline. Of these 7 patients, 1 in each arm had wild type virus genotype. Five patients in the saquinavir/r arm and 4 patients in the lopinavir/r arm failed with new reverse transcriptase (RT) mutations. One patient had new PI-associated mutations at virologic failure. These mutations developed shortly after a febrile illness with gastrointestinal symptoms that resulted in non-adherence with the medication regimen. Of the 16 virologic failures, 4/11 in the saquinavir/r arm and 3/5 in the lopinavir/r arm were documented as poorly adherent, based on 4-day patient recall of missed doses of >15% or as assessed by the investigator.
Time to loss of virologic response was analyzed based on response criteria of HIV-1 RNA <50 copies/mL and HIV-1 RNA <400 copies/mL. For patients who achieved a response of HIV-1 RNA <50 copies/mL, 16 (9.6%) in the saquinavir/r arm and 14 (8.2%) in the lopinavir/r arm lost virologic response. For patients who achieved a response of HIV-1 RNA <400 copies/mL, 23 (13.8%) in the saquinavir/r arm and 23 (13.5%) in the lopinavir/r arm lost virologic response. There were no statistically significant differences between the 2 treatment arms in the analysis of time to loss of virologic response using either HIV-1 RNA <50 copies/mL (P=0.4391) or HIV-1 RNA <400 copies/mL (P=0.5902).  The proportion of patients with hypertension was small in both treatment arms at all scheduled time points, ranging from 1.4% (2 patients in the saquinavir/r arm at week 36) to 6.0% (8 patients in the lopinavir/r arm at week 48). There was no trend for the proportion of patients with hypertension to increase or decrease over time, and no statistical significant treatment-arm difference was detected at any assessment.
Five (3%) ITT patients in the saquinavir/r treatment arm and 12 (7%) ITT patients in the lopinavir/r treatment arm discontinued study medication due to adverse events. In the lopinavir/r treatment arm, most of the events experienced by the 12 patients occurred before the week 12 visit, and 8 of the patients experienced gastrointestinal disorders.
Utilizing a 4-day drug recall questionnaire, more than 90% of patients in each treatment group reported at least 80% adherence at each study week.

Safety: During the study, 51% of patients in the saquinavir/r treatment arm and 63% in the lopinavir/r treatment arm reported adverse events. The most common AEs in the saquinavir/r arm were diarrhea (7% of patients), nausea (6%), vomiting (6%) and bronchitis (6%). The most common AEs in the lopinavir/r arm were diarrhea (14% of patients), nausea (9%) and vomiting (6%). The proportion of patients with gastrointestinal disorders was higher in the lopinavir/r arm than the saquinavir/r arm (27% vs. 17%), but the proportion of patients with AEs in other system organ classes was generally similar in the 2 treatment arms. Twenty-five (15%) patients in the saquinavir/r and 29 (17%) patients in the lopinavir/r arm had grade 3 or 4 AEs. Seven (4%) patients in the saquinavir/r arm and 8 (5%) patients in the lopinavir/r arm had AIDS-defining events during the study.
Three patients in the saquinavir/r arm and 4 patients in the lopinavir/r arm died during the study or follow-up period. One death in the saquinavir/r arm (a crime victim) was considered by the investigator to be remotely related to study treatment. The other 3 deaths were considered unrelated to study treatment. Twenty-four (15%) patients in the saquinavir/r arm had a total of 32 serious adverse events (SAEs) and 19 (11%) patients in the lopinavir/r arm had a total of 24 SAEs. The most common SAEs were infections and infestations (saquinavir/r arm, 7% of patients; lopinavir/r arm, 5% of patients). Five (3%) patients in the saquinavir/r arm and 12 (7%) patients in the lopinavir/r arm withdrew from study medication due to AEs. The most common AEs resulting in discontinuation of study medication were gastrointestinal disorders (saquinavir/r arm, 1 patient; lopinavir/r arm, 8 patients).
Mean and median changes from baseline in laboratory variables were generally small in both treatment arms with no apparent trend to increase or decrease during the 48-week treatment period. There were no statistically significant differences between treatment arms in the change from baseline in total cholesterol, LDL cholesterol, and HDL cholesterol. The change (elevation) from baseline in triglyceride levels was significantly lower in the saquinavir/r arm compared with the lopinavir/r arm at both week 24 (P=0.0007) and week 48 (P=0.0022). The change (decrease) from baseline in the total cholesterol: HDL ratio was significantly higher in the saquinavir/r arm than the lopinavir/r arm at week 24 (P=0.0237) but not at week 48 (P=0.4727).
An evaluation of selected treatment-emergent laboratory toxicities up to week 48 showed that triglyceride toxicity grade ≥2 and grade ≥3 occurred more frequently in the lopinavir/r arm (15 [9.6%] and 6 [3.8%] patients, respectively) than the saquinavir/r arm (1 [0.6%] and 1 [0.6%] patient, respectively). Grade ≥2 and grade ≥3 LDL cholesterol occurred more frequently in the saquinavir/r arm (23 [14.9%] and 6 [3.9%] patients, respectively) than the lopinavir/r arm (15 [9.6%] and 1 [0.6%] patients, respectively). Other apparent differences between treatment arms included grade ≥3 ALT and AST, which were more frequent in the lopinavir/r arm than the saquinavir/r arm, and grade ≥2 hemoglobin, which was also more frequent in the lopinavir/r arm than the saquinavir/r arm. Other laboratory toxicities were generally similar in the 2 treatment arms.

Conclusions
The saquinavir/r + emtricitabine/tenofovir treatment regimen was shown to be non-inferior to the lopinavir/r + emtricitabine/tenofovir treatment regimen in HIV-1 infected, treatment-naive adults. The extent of virologic suppression (decreased HIV-1 RNA viral load) and immunologic improvement (increased CD4+ lymphocyte count) was substantial and similar in the 2 treatment arms. Virologic failure was uncommon and similar between the 2 treatment arms. Emergence of major PI-associated resistance mutations occurred in a single patient in the context of confounding treatment issues.
The frequency of adverse events was similar in the 2 treatment arms, and few patients in either arm discontinued due to AEs. The elevation in triglyceride levels from baseline was significantly lower in the saquinavir/r arm compared with the lopinavir/r arm. There were no statistically significant differences between treatment arms in the change from baseline in total cholesterol, LDL cholesterol and HDL cholesterol.

Publications (references, if available)
Walmsley S, Ruxrungtham K, Slim J, Ward D, Larson P, Raffi F.
The Gemini Study: Saquinavir/r (SQV/r) vs lopinavir/r (LPV/r) plus emtricitabine/tenofovir (FTC/TDF) as initial therapy in HIV-1 infected patients. Presented at: 11th European AIDS Conference / EACS Madrid October 24-27, 2007.

Date of report
1/9/2008

Click here for the protocol registry listing of this trial.