The proportion of participants who achieved HIV-1 RNA <50 copies/mL at week 48 was assessed according to Food and Drug Administration's (FDA's) Missing, Switch, Discontinuation'=Failure (MSDF) Snapshot algorithm. The algorithm used the plasma HIV-1 RNA in the Week 48 visit window, followed the "virology-first principle" and considers a participant who has a missing plasma HIV-1 RNA, or switches to prohibited ARV regimen or discontinues from the study or study drug for any reason, or dies, as a failure.

The Full Analysis Set (FAS) consisted of all randomized participants who received at least one dose of the study drug. The missing value was imputed per FDA's MSDF Snapshot algorithm as described under "Outcome Measure Description" above.

Number of participants with treatment-emergent non serious AEs

Safety Analysis Set was the same as the FAS consisting of all randomized participants who received at least one dose of the study drug but analyzed as realized for tropism assay and treatment.

Number of participants with grade 3 or 4 AEs are presented here.

Safety Analysis Set was the same as the FAS consisting of all randomized participants who received at least one dose of the study drug but analyzed as realized for tropism assay and treatment.

Number of participants who discontinued due to AEs are reported here. Three participants (two from the MVC+DRV/r arm and one from the FTC/TDF+DRV/r arm) were not considered as discontinued due to AE because other reasons for discontinuation were prioritized for these participants.

Safety Analysis Set was the same as the FAS consisting of all randomized participants who received at least one dose of the study drug but analyzed as realized for tropism assay and treatment.

Number of treatment-related AEs are presented here.

Safety Analysis Set was the same as the FAS consisting of all randomized participants who received at least one dose of the study drug but analyzed as realized for tropism assay and treatment.

Total number of participants with treatment-emergent serious adverse events are reported

Safety Analysis Set was the same as the FAS consisting of all randomized participants who received at least one dose of the study drug but analyzed as realized for tropism assay and treatment.

Number of participants with laboratory abnormalities are reported

Safety Analysis Set was the same as the FAS consisting of all randomized participants who received at least one dose of the study drug but analyzed as realized for tropism assay and treatment. One participant was not analyzed for laboratory data as the collection date for all lab data was less than the first active therapy date.

Number of participants who had clinically significant laboratory abnormalities of Grade 3 and Grade 4 according to DAIDS. Abnormality incidence of highest grade was reported for a labcode for each individual participant.

Safety Analysis Set was the same as the FAS consisting of all randomized participants who received at least one dose of the study drug but analyzed as realized for tropism assay and treatment.

The relationship of the proportion of participants achieving HIV-1 RNA <50 copies/mL at Week 48 with the screening tropism test for the MVC containing regimen was analyzed. Virologic response for a participant at Week 48 was derived using the FDA's Snapshot MSDF algorithm. Difference in proportions of patients with plasma HIV-1 RNA <50 copies/mL at week 48 between the maraviroc and the emtricitabine/tenofovir treatment arms, with two-sided 95% confidence interval, among patients who are R5 by genotype (including some who were originally randomized to ESTA and are R5 by genotype upon retesting), were calculated via the Maximum Likelihood method. The estimate was adjusted for the screening plasma HIV RNA level (<100,000 vs. ≥100,000) copies/mL via the Mantel Haenszel (MH) method.

The FAS consisted of all randomized participants who received at least one dose of the study drug.

Per the protocol participants who meet the following criteria were regarded as PDTFs requiring a confirmatory plasma HIV-1 RNA determination: • Decrease in plasma HIV-1 RNA <1 log10 from baseline after Week 4 unless plasma HIV-1 RNA is <50 copies/mL, or • Plasma HIV-1 RNA >1.0 log10 above the nadir value after Week 4 where the nadir is the lowest plasma HIV-1 RNA concentration, or • Plasma HIV-1 RNA ≥50 copies/mL at any time after Week 24, or • Plasma HIV-1 RNA ≥50 copies/mL after suppression to <50 copies/mL on two consecutive visits, or • Decrease in plasma HIV-1 RNA ≤2 log10 from baseline on or after Week 12 unless plasma HIV-1 RNA is <400 copies/mL. Decrease in plasma HIV-1 RNA ≤2 log10 from baseline on or after Week 12 unless plasma HIV-1 RNA is <50 copies/mL (before August 30 2012) or <400 copies/mL (after August 30 2012).

The FAS consisted of all randomized participants who received at least one dose of the study drug. No imputation for missing values was performed for this endpoint. 'Evaluability' is determined by the on-treatment viral load (≥400 copies/mL at sample time point).

For participants meeting the PDTF criteria, tropism was assessed using the original randomized and alternate assays (ie, both genotype testing and ESTA). Data reported here corresponds to the timepoint at or after PDTF.

Number of Evaluable PDTF = Virology Analysis Population (VAP) 'Evaluability' is determined by the on-treatment viral load (≥400 copies/mL at sample time point).

For participants meeting the PDTF criteria, viral resistance to maraviroc for maraviroc treated participants was assessed in patients with R5 virus at failure. The resistance level is calculated by reference to a laboratory strain of virus that is analyzed in parallel with the clinical isolate to identify 50% inhibitory concentrations (IC50). The maximal percent inhibition is the percent inhibition that is achieved in a titration of the drug at high concentrations when the addition of more drug does not result in increased inhibition. Maximal percent inhibition is obtained in the same way as the titration for IC50, but the key measure is of the plateau height of percent inhibition, where increased concentration of maraviroc does not result in additional inhibition. This is consistent with the virus developing some ability to use maraviroc-bound CCR5 for entry. A significant change in IC50 is not required for this mechanism.

The FAS consisted of all randomized participants who received at least one dose of the study drug.

For participants meeting the PDTF criteria, viral resistance (both genotypic and phenotypic) to NRTI, NNRTI, and PI's were assessed at Baseline and on-treatment. The assessment was performed using the overall (i.e. net) susceptibility score provided using the PhenoSense GT assay. The number of participants with successful assessments were 15/17 for the MVC+DRV/r arm and 3/3 for the FTC/TDF+DRV/r arm.

The FAS consisted of all randomized participants who received at least one dose of the study drug. The assessment was performed using the overall (ie. net) susceptibility score provided using the PhenoSense GT assay. The number of participants with successful assessments were 15 for the MVC+DRV/r arm and 3 for the FTC/TDF+DRV/r arm.

The differences in the magnitude of changes in CD4+ at Baseline and at Week 48 for maraviroc versus emtricitabine/tenofovir were compared.

The FAS consisted of all randomized participants who received at least one dose of the study drug. Missing values were imputed using LOCF approach. Baseline was calculated as the average of all the pre-dose measurements excluding the screening value. If all pre-dose values were missing, screening value was considered as the baseline.

The differences in the magnitude of changes in CD4+ from Baseline through Week 48 for maraviroc versus emtricitabine/tenofovir were compared.

The FAS consisted of all randomized participants who received at least one dose of the study drug. Missing values were imputed using LOCF approach. Baseline was calculated as the average of all the pre-dose measurements excluding the screening value. If all pre-dose values were missing, screening value was considered as the baseline.

The differences in the magnitude of changes in CD8+ cell counts from baseline through Week 48 for maraviroc versus emtricitabine/tenofovir were compared.

The FAS consisted of all randomized participants who received at least one dose of the study drug. Missing values were imputed using LOCF approach. Baseline was calculated as the average of all the pre-dose measurements excluding the screening value. If all pre-dose values were missing, screening value was considered as the baseline.

The differences in the magnitude of changes in CD8+ cell counts from Baseline through Week 48 for maraviroc versus emtricitabine/tenofovir were compared.

The FAS consisted of all randomized participants who received at least one dose of the study drug. Missing values were imputed using LOCF approach. Baseline was calculated as the average of all the pre-dose measurements excluding the screening value. If all pre-dose values were missing, screening value was considered as the baseline.

The differences in the magnitude of changes in CD4+/CD8+ ratio from Baseline through Weeks 48 for maraviroc versus emtricitabine/tenofovir were compared.

The FAS consisted of all randomized participants who received at least one dose of the study drug. Missing values were imputed using LOCF approach. Baseline was calculated as the average of all the pre-dose measurements excluding the screening value. If all pre-dose values were missing, screening value was considered as the baseline.

A sub-study was conducted in which the participants underwent whole-body DEXA scans to evaluate peripheral fat tissue estimates for left and right arms and legs and truncal fat mass and truncal lean mass. Truncal abdominal fat were estimated from the DEXA scan field set on the torso. The effects on estimates of fat mass and lean mass were addressed by providing LSMs of change from baseline.

Safety Analysis Set was the same as the FAS consisting of all randomized participants who received at least one dose of the study drug but analyzed as realized for tropism assay and treatment. Missing values were imputed using LOCF approach.

A sub-study was conducted in which the participants underwent whole-body DEXA scans to evaluate peripheral fat tissue estimates for left and right arms and legs and truncal fat mass and truncal lean mass. Truncal abdominal fat were estimated from the DEXA scan field set on the torso. The effects on estimates of fat mass and lean mass were addressed by providing LSMs of change from baseline.

Safety Analysis Set was the same as the FAS consisting of all randomized participants who received at least one dose of the study drug but analyzed as realized for tropism assay and treatment. Missing values were imputed using LOCF approach.

Bone mineral density was evaluated by DEXA scan in a subset of participants who consented to these evaluations. The effects on BMD were addressed by providing LSMs of change from baseline bone mineral density of the lumbar spine (L1-L4), left total hip and femoral neck as measured by the DEXA scan.

Safety Analysis Set was the same as the FAS consisting of all randomized participants who received at least one dose of the study drug but analyzed as realized for tropism assay and treatment. Missing values were imputed using LOCF approach.

Bone mineral density was evaluated by DEXA scan in a subset of participants who consented to these evaluations. The effects on BMD were addressed by providing LSMs of change from Baseline bone mineral density femoral neck as measured by the DEXA scan.

Safety Analysis Set was the same as the FAS consisting of all randomized participants who received at least one dose of the study drug but analyzed as realized for tropism assay and treatment. Missing values were imputed using LOCF approach.

Bone mineral density was evaluated by DEXA scan in a subset of participants who consented to these evaluations. The effects on BMD were addressed by providing LSMs of change from baseline bone mineral density of the lumbar spine (L1-L4) as measured by the DEXA scan.

Safety Analysis Set was the same as the FAS consisting of all randomized participants who received at least one dose of the study drug but analyzed as realized for tropism assay and treatment. Missing values were imputed using LOCF approach.

Bone turnover marker, osteocalcin, was collected in the subset of participants participating in the DEXA scan sub-study.

Safety Analysis Set was the same as the FAS consisting of all randomized participants who received at least one dose of the study drug but analyzed as realized for tropism assay and treatment. Missing values were imputed using LOCF approach.

Bone turnover marker, C-telopeptide of type 1 collagen (CTx), was collected in the subset of participants participating in the DEXA scan sub-study.

Safety Analysis Set was the same as the FAS consisting of all randomized participants who received at least one dose of the study drug but analyzed as realized for tropism assay and treatment. Missing values were imputed using LOCF approach.