A Study of Trastuzumab Emtansine (T-DM1) Plus Pertuzumab/Pertuzumab Placebo Versus Trastuzumab [Herceptin] Plus a Taxane in Participants With Metastatic Breast Cancer (MARIANNE)

Tumor assessments were performed according to Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1, using radiographic images submitted to the IRF up to and including the confirmatory tumor assessment 4 to 6 weeks after study drug discontinuation. Disease progression was defined as a greater than or equal to (≥) 20 percent (%) and 5-millimeter (mm) increase in sum of diameters of target lesions, taking as reference the smallest sum obtained during the study, or appearance of new lesion(s). The percentage of participants with death or disease progression was calculated as \[number of participants with event divided by the number analyzed\] multiplied by 100.

Time frame: Up to 48 months from randomization until clinical cutoff of 16-Sept-2014 (at Screening, every 9 weeks for 81 weeks, then every 12 weeks thereafter and/or up to 42 days after last dose)

Population: ITT Population.

Tumor assessments were performed according to RECIST version 1.1, using radiographic images submitted to the IRF up to and including the confirmatory tumor assessment 4 to 6 weeks after study drug discontinuation. PFS was defined as the time from randomization to first documented disease progression or death from any cause. Disease progression was defined as a ≥20% and 5-mm increase in sum of diameters of target lesions, taking as reference the smallest sum obtained during the study, or appearance of new lesion(s). Median duration of PFS was estimated using Kaplan-Meier analysis, and corresponding confidence intervals (CIs) were computed using the Brookmeyer-Crowley method.

Time frame: Up to 48 months from randomization until clinical cutoff of 16-Sept-2014 (at Screening, every 9 weeks for 81 weeks, then every 12 weeks thereafter and/or up to 42 days after last dose)

Population: ITT Population.

The WPAI is a patient-reported measure which assesses the effect of general health and symptom severity on work productivity and regular activities. The General Health questionnaire asks participants to estimate the number of hours missed from work due to reasons related and unrelated to their health problems, as well as the total number of hours actually worked in the preceding 7-day period. The percentage of participants reporting that they were employed (working for pay) was assessed at baseline was assessed along with Absenteeism (work time missed), Presenteeism (impairment at work / reduced on-the-job effectiveness), Work productivity loss (overall work impairment / absenteeism plus presenteeism), and Activity Impairment. The reported changes represent change from baseline at Cycle 7. The score range for the scales of the WPAI is between 0 (no effect) to 100% (max effect).

Time frame: Baseline, Cycle 7 (Week 18)

Population: Number of participants analysed=participants from ITT population who reported conduct of daily activities. Here, 'n' signifies the number of participants with available data at specified category.

The RSCL is a self-reported instrument which consists of 4 domains including physical symptom distress, psychological distress, activity level, and overall global life quality. Only the activity level scale was collected and assessed. Scores may range from 0 to 100, with higher scores indicating increased burden of disease. Mean RSCL activity scale score changes were calculated as \[mean score at the assessment visit minus mean score at Baseline\]. The higher the score, the higher the level of impairment or burden.

Time frame: Baseline, Cycle 7 (Week 18)

Population: ITT Population. Here, 'n' signifies the number of participants with available data at baseline and Cycle 7 (Week 18).

The WPAI is a patient-reported measure which assesses the effect of general health and symptom severity on work productivity and regular activities. The General Health questionnaire asks participants to estimate the number of hours missed from work due to reasons related and unrelated to their health problems, as well as the total number of hours actually worked in the preceding 7-day period. The percentage of participants reporting that they were employed (working for pay) was assessed at baseline was assessed along with Absenteeism (work time missed), Presenteeism (impairment at work / reduced on-the-job effectiveness), Work productivity loss (overall work impairment / absenteeism plus presenteeism), and Activity Impairment. The reported changes represent change from baseline at Cycle 7. The score range for the scales of the WPAI is between 0 (no effect) to 100% (max effect)

Time frame: Baseline, Cycle 7 (Week 18)

Population: Number of participants analysed=participants from ITT population who were employed at baseline. Here, 'n' signifies the number of participants with available data at specified category.

Tumor assessments were performed according to RECIST version 1.1, using radiographic images submitted to the IRF up to and including the confirmatory tumor assessment 4 to 6 weeks after study drug discontinuation. Duration of response was defined as the time from confirmed PR or CR to first documented disease progression or death from any cause. CR was defined as the disappearance of all target lesions and non-target lesions and short-axis reduction in pathological lymph nodes to \<10 mm. PR was defined as a ≥30% decrease in sum of diameters of target lesions, taking as reference the Baseline sum. Response was determined using 2 consecutive tumor assessments at least 4 weeks apart. Median duration of response was estimated using Kaplan-Meier analysis, and corresponding CIs were computed using the Brookmeyer-Crowley method.

Time frame: Up to 46 months from randomization until clinical cutoff of 16-Sept-2014 (at Screening, every 9 weeks for 81 weeks, then every 12 weeks thereafter and/or up to 42 days after last dose)

Population: ITT Population. Only participants achieving CR or PR were included in the analysis.

Hospitalization was defined as a non-administration-related hospitalization due to serious adverse event, while on study treatment. Reported values represent number of days admitted per participants.

Time frame: Up to 48 months from randomization until clinical cutoff of 16-Sept-2014

Population: Safety population. Number of participants analyzed=participants with hospitalization and data available for calculation of the parameter.

The percentage of participants alive at 1 year after randomization was estimated as the one-year survival rate using Kaplan-Meier analysis, and corresponding CIs were computed using Greenwood's estimate of the standard error.

Time frame: From randomization until 1 year

Population: ITT Population.

OS was defined as the time from randomization to death from any cause. Median duration of OS was estimated using Kaplan-Meier analysis.

Time frame: Up to 70 months from randomization until clinical cutoff of 15-May-2016 (every 3 months until death, loss to follow-up, withdrawal, or study termination)

Population: ITT Population (High HER2 mRNA Subpopulation).

OS was defined as the time from randomization to death from any cause. Median duration of OS was estimated using Kaplan-Meier analysis. Reported upper bound of confidence interval for Trastuzumab Emtansine + Placebo and confidence interval values for Trastuzumab + Taxane and Trastuzumab Emtansine + Pertuzumab are censored values.

Time frame: Up to 70 months from randomization until clinical cutoff of 15-May-2016 (every 3 months until death, loss to follow-up, withdrawal, or study termination)

Population: ITT Population (Low HER2 mRNA Subpopulation).

OS was defined as the time from randomization to death from any cause. Median duration of OS was estimated using Kaplan-Meier analysis, and corresponding CIs were computed using the Brookmeyer-Crowley method.

Time frame: Up to 70 months from randomization until clinical cutoff of 15-May-2016 (every 3 months until death, loss to follow-up, withdrawal, or study termination)

Population: ITT Population.

Overall Survival truncated at 2 years was defined as the percentage of participants alive at 2 years.

Time frame: From randomization until 2 years

Population: ITT Population.

The FACT-Taxane is a self-reported instrument which measures the health-related quality of life (HRQOL) of participants receiving taxane-containing chemotherapy. The FACT-TaxS consists of 16 items including 11 neurotoxicity-related questions and 5 additional questions assessing arthralgia, myalgia, and skin discoloration. Items are rated from 0 (not at all) to 4 (very much) and a total score is inversely derived. Scores may range from 0 to 64, with higher scores indicating fewer/no symptoms. A minimally clinically important difference in treatment-related symptoms was defined as a ≥5% decrease (ie, 3.2 points) in FACT-TaxS score from Baseline. The percentage of participants with treatment-related symptoms was calculated using following formula: \[number of participants meeting the above threshold divided by the number analyzed\] multiplied by 100. Corresponding CIs were computed using the Blyth-Still-Casella exact method.

Time frame: Up to 39 months from randomization until clinical cutoff of 16-Sept-2014 (at Baseline, on Day 1 of Cycles 2 to 8 and every even-numbered cycle thereafter and/or up to 42 days after last dose)

Population: ITT Population (Protocol Amendment C Subpopulation): All randomized participants who entered the study after Protocol Amendment C.

Treatment failure was defined as the discontinuation of all study medications in the treatment arm for any reason including disease progression, treatment toxicity, death, physician decision, or participant withdrawal. The percentage of participants with treatment failure was calculated as \[number of participants with event divided by the number analyzed\] multiplied by 100.

Time frame: Up to 48 months from randomization until clinical cutoff of 16-Sept-2014

Population: ITT Population.

The FACT-C is a self-reported instrument which measures HRQOL pertaining to colorectal cancer. Response options on each question may range from 0 (not at all) to 4 (very much). The percentage of participants with diarrhea was calculated using following formula: \[number of participants with any level of either symptom divided by the number analyzed\] multiplied by 100.

Time frame: At Baseline, Day 8 of Cycle 1, and Days 1 and 8 of Cycle 2

Population: ITT Population (Protocol Amendment C Subpopulation). Only participants with a FACT-C score at the designated visit (n) were included in the analysis.

The FACT-C is a self-reported instrument which measures HRQOL pertaining to colorectal cancer. Response options on each question may range from 0 (not at all) to 4 (very much). The percentage of participants with nausea was calculated using following formula: \[number of participants with any level of either symptom divided by the number analyzed\] multiplied by 100.

Time frame: At Baseline, Day 8 of Cycle 1, and Days 1 and 8 of Cycle 2

Population: ITT Population (Protocol Amendment C Subpopulation). Only participants with a FACT-C score at the designated visit (n) were included in the analysis.

Time frame: From randomization until 2 years

Population: ITT Population

The percentage of participants who died prior to clinical cutoff was calculated as \[number of participants with event divided by the number analyzed\] multiplied by 100.

Time frame: Up to 70 months from randomization until clinical cutoff of 15-May-2016 (every 3 months until death, loss to follow-up, withdrawal, or study termination)

Population: ITT Population.

The percentage of participants who died prior to clinical cutoff was calculated as \[number of participants with event divided by the number analyzed\] multiplied by 100.

Time frame: Up to 70 months from randomization until clinical cutoff of 15-May-2016 (every 3 months until death, loss to follow-up, withdrawal, or study termination)

Population: ITT Population (High HER2 mRNA Subpopulation).

The percentage of participants who died prior to clinical cutoff was calculated as \[number of participants with event divided by the number analyzed\] multiplied by 100.

Time frame: Up to 70 months from randomization until clinical cutoff of 15-May-2016 (every 3 months until death, loss to follow-up, withdrawal, or study termination)

Population: ITT Population (Low HER2 mRNA Subpopulation).

Tumor assessments were performed according to RECIST version 1.1, using radiographic images submitted to the IRF up to and including the confirmatory tumor assessment 4 to 6 weeks after study drug discontinuation. CR was defined as the disappearance of all target and non-target lesions and short-axis reduction in pathological lymph nodes to \<10 mm. PR was defined as a ≥30% decrease in sum of diameters of target lesions, taking as reference the Baseline sum. Response was determined using 2 consecutive tumor assessments at least 4 weeks apart. The percentage of participants with a best overall response of CR or PR (ie, the ORR) was calculated as \[number of participants meeting the respective criteria divided by the number analyzed\] multiplied by 100.

Time frame: Up to 46 months from randomization until clinical cutoff of 16-Sept-2014 (at Screening, every 9 weeks for 81 weeks, then every 12 weeks thereafter and/or up to 42 days after last dose)

Population: ITT Population (High HER2 mRNA Subpopulation): All randomized participants with above-the-median HER2 mRNA expression (value greater than \[\>\] 59.71). Only participants with measurable disease at Baseline were included in the analysis.

Tumor assessments were performed according to RECIST version 1.1, using radiographic images submitted to the IRF up to and including the confirmatory tumor assessment 4 to 6 weeks after study drug discontinuation. CR was defined as the disappearance of all target and non-target lesions and short-axis reduction in pathological lymph nodes to \<10 mm. PR was defined as a ≥30% decrease in sum of diameters of target lesions, taking as reference the Baseline sum. Response was determined using 2 consecutive tumor assessments at least 4 weeks apart. The percentage of participants with a best overall response of CR or PR (ie, the ORR) was calculated as \[number of participants meeting the respective criteria divided by the number analyzed\] multiplied by 100.

Time frame: Up to 46 months from randomization until clinical cutoff of 16-Sept-2014 (at Screening, every 9 weeks for 81 weeks, then every 12 weeks thereafter and/or up to 42 days after last dose)

Population: ITT Population (Low HER2 mRNA Subpopulation): All randomized participants with below-the-median HER2 mRNA expression (value less than or equal to \[≤\] 59.71). Only participants with measurable disease at Baseline were included in the analysis.

Tumor assessments were performed according to RECIST version 1.1, using radiographic images submitted to the IRF up to and including the confirmatory tumor assessment 4 to 6 weeks after study drug discontinuation. CR was defined as the disappearance of all target and non-target lesions and short-axis reduction in pathological lymph nodes to \<10 mm. PR was defined as a ≥30% decrease in sum of diameters of target lesions, taking as reference the Baseline sum. SD was defined as neither sufficient shrinkage to quality for PR nor sufficient (20%) increase to qualify for disease progression. Response was determined using 2 consecutive tumor assessments at least 4 weeks apart. The percentage of participants with a best overall response of CR, PR, or SD was calculated as \[number of participants meeting the respective criteria divided by the number analyzed\] multiplied by 100.

Time frame: Up to 46 months from randomization until clinical cutoff of 16-Sept-2014 (at Screening, every 9 weeks for 81 weeks, then every 12 weeks thereafter and/or up to 42 days after last dose)

Population: Data were not analyzed. Protocol Amendment E removed this outcome measure as a secondary endpoint, because it was redundant to another prespecified secondary endpoint.

The FACT-B is a self-reported instrument which measures HRQOL of participants with breast cancer. It consists of 5 subscales including physical well-being (PWB), social well-being (SWB), emotional well-being (EWB), functional well-being (FWB), and a breast cancer subscale (BCS). The TOI-PFB score is taken by adding the scores from the PWB (7 items), FWB (7 items), and BCS (9 items) subscales. Items are rated from 0 (not at all) to 4 (very much) and a total score is derived. Scores may range from 0 to 92, with higher scores indicating better HRQOL. A 5 point change has been identified as the clinically minimal important difference (CMID) on the FACT-TOI-PFB scale. The percentage of participants with deterioration was calculated as \[number of participants meeting the above threshold divided by the number analyzed\] multiplied by 100.

Time frame: Up to 39 months from randomization until clinical cutoff of 16-Sept-2014 (at Baseline, on Day 1 of Cycles 2 to 8 and every even-numbered cycle thereafter and/or up to 42 days after last dose)

Population: ITT Population. Number of participants analyzed=participants with baseline and at least one post baseline FACT-B TOI-PFB score.

Tumor assessments were performed by the investigator according to RECIST version 1.1. Disease progression was defined as a ≥20% and 5-mm increase in sum of diameters of target lesions, taking as reference the smallest sum obtained during the study, or appearance of new lesion(s). The percentage of participants with death or disease progression was calculated as \[number of participants with event divided by the number analyzed\] multiplied by 100.

Time frame: Up to 48 months from randomization until clinical cutoff of 16-Sept-2014 (at Screening, every 9 weeks for 81 weeks, then every 12 weeks thereafter and/or up to 42 days after last dose)

Population: ITT Population.

Tumor assessments were performed according to RECIST version 1.1, using radiographic images submitted to the IRF up to and including the confirmatory tumor assessment 4 to 6 weeks after study drug discontinuation. Disease progression was defined as a ≥20% and 5-mm increase in sum of diameters of target lesions, taking as reference the smallest sum obtained during the study, or appearance of new lesion(s). The percentage of participants with death or disease progression was calculated as \[number of participants with event divided by the number analyzed\] multiplied by 100.

Time frame: Up to 48 months from randomization until clinical cutoff of 16-Sept-2014 (at Screening, every 9 weeks for 81 weeks, then every 12 weeks thereafter and/or up to 42 days after last dose)

Population: ITT Population (High HER2 mRNA Subpopulation).

Tumor assessments were performed according to RECIST version 1.1, using radiographic images submitted to the IRF up to and including the confirmatory tumor assessment 4 to 6 weeks after study drug discontinuation. Disease progression was defined as a ≥20% and 5-mm increase in sum of diameters of target lesions, taking as reference the smallest sum obtained during the study, or appearance of new lesion(s). The percentage of participants with death or disease progression was calculated as \[number of participants with event divided by the number analyzed\] multiplied by 100.

Time frame: Up to 48 months from randomization until clinical cutoff of 16-Sept-2014 (at Screening, every 9 weeks for 81 weeks, then every 12 weeks thereafter and/or up to 42 days after last dose)

Population: ITT Population (Low HER2 mRNA Subpopulation).

The ECOG performance status is a scale used to quantify cancer participants' general well-being and activities of daily life. The scale ranges from 0 to 5, with 0 denoting perfect health and 5 indicating death. The 6 categories are 0=Asymptomatic (Fully active, able to carry on all predisease activities without restriction), 1=Symptomatic but completely ambulatory (Restricted in physically strenuous activity but ambulatory and able to carry out work of a light or sedentary nature), 2=Symptomatic, \< 50% in bed during the day (Ambulatory and capable of all self-care but unable to carry out any work activities. Up and about more than 50% of waking hours), 3=Symptomatic, \> 50% in bed, but not bedbound (Capable of only limited self-care, confined to bed or chair 50% or more of waking hours), 4=Bedbound (Completely disabled. Cannot carry on any self-care. Totally confined to bed or chair), 5=Death.

Time frame: Baseline, Day 1 of every Cycle up to Clinical Data Cut (up to 48 months)

Population: Safety population

Laboratory results were graded according to NCI CTCAE version 4.0. Grade 3: Severe or medically significant but not immediately life-threatening; hospitalization or prolongation of hospitalization indicated; disabling; limiting self care activity of daily living with inability to perform bathing, dressing and undressing, feeding self, using the toilet, taking medications but not bedridden. Grade 4: An immediate threat to life. Urgent medical intervention is required in order to maintain survival.

Time frame: Day 1, 8, and 15 of Cycle 1-3 and on Day 1 of each subsequent cycle up to 50 months from randomization until clinical cutoff of 16-Sept-2014

Population: Safety population. Number of participants analyzed=participants with available data for the outcome.

Adverse events were graded according to National Cancer Institute (NCI) Common Terminology Criteria for Adverse Events (CTCAE) version 4.0. Grade 3: Severe or medically significant but not immediately life-threatening; hospitalization or prolongation of hospitalization indicated; disabling; limiting self care activity of daily living with inability to perform bathing, dressing and undressing, feeding self, using the toilet, taking medications but not bedridden. Grade 4: An immediate threat to life. Urgent medical intervention is required in order to maintain survival. Grade 5: Death.

Time frame: Up to 50 months from randomization until clinical cutoff of 16-Sept-2014 (continuously until 28 days after last dose

Population: Safety Population: All treated participants. Additionally, 2 participants randomized to trastuzumab+taxane received 3 cycles of trastuzumab emtansine and were included in trastuzumab emtansine+placebo arm. 6 participants randomized to trastuzumab emtansine+placebo received pertuzumab and were included in trastuzumab emtansine+pertuzumab arm.

Adverse events were graded according to NCI CTCAE version 4.0. Grade 5 adverse events are those events which led to death.

Time frame: Up to 50 months from randomization until clinical cutoff of 16-Sept-2014 (continuously until 28 days after last dose)

Population: Safety population

Hospitalization was defined as a non-administration-related hospitalization due to serious adverse event, while on study treatment.

Time frame: Up to 48 months from randomization until clinical cutoff of 16-Sept-2014

Population: Safety population

Objective response was defined as having CR or PR, assessed according to RECIST version 1.1, by investigator. CR was defined as the disappearance of all target and non-target lesions and short-axis reduction in pathological lymph nodes to \<10 mm. PR was defined as a ≥30% decrease in sum of diameters of target lesions, taking as reference the Baseline sum. Response was determined using 2 consecutive tumor assessments at least 4 weeks apart. The percentage of participants with a best overall response of CR or PR (ie, the ORR) was calculated as \[number of participants meeting the respective criteria divided by the number analyzed\] multiplied by 100. Corresponding CIs were computed using the Blyth-Still-Casella exact method.

Time frame: Up to 46 months from randomization until clinical cutoff of 16-Sept-2014 (at Screening, every 9 weeks for 81 weeks, then every 12 weeks thereafter and/or up to 42 days after last dose)

Population: ITT Population.

Objective response was defined as having complete response (CR) or partial response (PR), assessed according to RECIST version 1.1, using radiographic images submitted to the IRF up to and including the confirmatory tumor assessment 4 to 6 weeks after study drug discontinuation. CR was defined as the disappearance of all target and non-target lesions and short-axis reduction in pathological lymph nodes to \<10 mm. PR was defined as a ≥30% decrease in sum of diameters of target lesions, taking as reference the Baseline sum. Response was determined using 2 consecutive tumor assessments at least 4 weeks apart. The percentage of participants with a best overall response of CR or PR (ie, the objective response rate \[ORR\]) was calculated as \[number of participants meeting the respective criteria divided by the number analyzed\] multiplied by 100. Corresponding CIs were computed using the Blyth-Still-Casella exact method.

Time frame: Up to 46 months from randomization until clinical cutoff of 16-Sept-2014 (at Screening, every 9 weeks for 81 weeks, then every 12 weeks thereafter and/or up to 42 days after last dose)

Population: ITT Population. Only participants with measurable disease at Baseline were included in the analysis.

Tumor assessments were performed by the investigator according to RECIST version 1.1. PFS was defined as the time from randomization to first documented disease progression or death from any cause. Disease progression was defined as a ≥20% and 5-mm increase in sum of diameters of target lesions, taking as reference the smallest sum obtained during the study, or appearance of new lesion(s). Median duration of PFS was estimated using Kaplan-Meier analysis, and corresponding CIs were computed using the Brookmeyer-Crowley method.

Time frame: Up to 48 months from randomization until clinical cutoff of 16-Sept-2014 (at Screening, every 9 weeks for 81 weeks, then every 12 weeks thereafter and/or up to 42 days after last dose)

Population: ITT Population.

Tumor assessments were performed according to RECIST version 1.1, using radiographic images submitted to the IRF up to and including the confirmatory tumor assessment 4 to 6 weeks after study drug discontinuation. PFS was defined as the time from randomization to first documented disease progression or death from any cause. Disease progression was defined as a ≥20% and 5-mm increase in sum of diameters of target lesions, taking as reference the smallest sum obtained during the study, or appearance of new lesion(s). Median duration of PFS was estimated using Kaplan-Meier analysis.

Time frame: Up to 48 months from randomization until clinical cutoff of 16-Sept-2014 (at Screening, every 9 weeks for 81 weeks, then every 12 weeks thereafter and/or up to 42 days after last dose)

Population: ITT Population (High HER2 mRNA Subpopulation).

Tumor assessments were performed according to RECIST version 1.1, using radiographic images submitted to the IRF up to and including the confirmatory tumor assessment 4 to 6 weeks after study drug discontinuation. PFS was defined as the time from randomization to first documented disease progression or death from any cause. Disease progression was defined as a ≥20% and 5-mm increase in sum of diameters of target lesions, taking as reference the smallest sum obtained during the study, or appearance of new lesion(s). Median duration of PFS was estimated using Kaplan-Meier analysis.

Time frame: Up to 48 months from randomization until clinical cutoff of 16-Sept-2014 (at Screening, every 9 weeks for 81 weeks, then every 12 weeks thereafter and/or up to 42 days after last dose)

Population: ITT Population (Low HER2 mRNA Subpopulation).

The FACT-B is a self-reported instrument which measures HRQOL of participants with breast cancer. It consists of 5 subscales including PWB, SWB, EWB, FWB, and BCS. The TOI-PFB score is taken by adding the scores from the PWB (7 items), FWB (7 items), and BCS (9 items) subscales. Items are rated from 0 (not at all) to 4 (very much) and a total score is derived. Scores may range from 0 to 92, with higher scores indicating better HRQOL. A 5 point change has been identified as the clinically minimal important difference (CMID) on the FACT-TOI-PFB scale. Time to deterioration was defined as the time from Baseline until the first decrease in FACT-B TOI-PFB score. Median time to deterioration was estimated using Kaplan-Meier analysis, and corresponding CIs were computed using the Brookmeyer-Crowley method.

Time frame: Baseline up to 39 months from randomization until clinical cutoff of 16-Sept-2014

Population: ITT Population. Number of participants analyzed=participants with baseline and at least one post baseline FACT-B TOI-PFB score.

Treatment failure was defined as the discontinuation of all study medications in the treatment arm for any reason including disease progression, treatment toxicity, death, physician decision, or participant withdrawal. TTF was defined as the time from randomization to treatment failure. Median TTF was estimated using Kaplan-Meier analysis, and corresponding CIs were computed using the Brookmeyer-Crowley method.

Time frame: Up to 48 months from randomization until clinical cutoff of 16-Sept-2014

Population: ITT Population.