Relapse defined as meeting any of these criteria: 50% increase in Montgomery-Asberg Depression Rating Scale (MADRS) score from randomization with a Clinical Global Impressions-Severity (CGI-S) of Depression score increase to a score of 4 or more; Hospitalized for depression or suicidality; Discontinued due to lack of efficacy/worsening of depression/suicidality. MADRS is a 10-item rating scale for depressive mood symptoms severity, items rated on 0-6 scale, with total score range of 0 (low severity of depressive symptoms) to 60 (high severity of depressive symptoms). CGI-S measures severity of illness at time of assessment compared with start of treatment. Scores range from 1 (normal, not at all ill) to 7 (the most extremely ill). Lack of Efficacy/Worsening of depression was at discretion of investigator based on clinical observation. Suicidality is thoughts or actions of self-harm as determined by the investigator. Those who did not relapse were "censored" at their last observation.

All randomized participants are included in the time to event analyses. The numbers of participants censored are 186 and 152 for OFC and Flu groups, respectively.

Relapse is defined as meeting any of the following criteria: 50% increase in Montgomery-Asberg Depression Rating Scale (MADRS) score from randomization with concomitant Clinical Global Impressions-Severity (CGI-S) of Depression score increase to a score of 4 or more; Hospitalization for depression or suicidality; Discontinuation due to lack of efficacy/worsening of depression/suicidality. MADRS is a rating scale for severity of depressive mood symptoms with 10 items rated on a scale of 0-6, for a total score range of 0 (low severity of depressive symptoms) to 60 (high severity of depressive symptoms). CGI-S measures severity of illness at the time of assessment compared with start of treatment. Scores range from 1 (normal, not at all ill) to 7 (the most extremely ill). Lack of Efficacy/Worsening of depression was at the discretion of the investigator and based on clinical observation. Suicidality is thoughts or actions of self-harm as determined by the investigator.

All randomized participants.

Relapse is defined as a 50% increase in the Montgomery-Asberg Depression Rating Scale (MADRS) score from randomization with concomitant Clinical Global Impressions-Severity (CGI-S) of Depression score increase to a score of 4 or more. The MADRS has a 10-item checklist with items rated on a scale of 0-6, for a total score range of 0 (low severity of depressive symptoms) to 60 (high severity of depressive symptoms). CGI-S measures severity of illness at the time of assessment compared with start of treatment. Scores range from 1 (normal, not at all ill) to 7 (the most extremely ill).

All randomized participants.

All randomized participants.

Lack of Efficacy/Worsening of depression was at the discretion of the investigator and was based on clinical observation. Suicidality is thoughts or actions of self-harm as determined by the investigator.

All randomized participants.

Relapse is defined as a 50% increase in the Montgomery-Asberg Depression Rating Scale (MADRS) score from randomization with concomitant Clinical Global Impressions-Severity (CGI-S) of Depression score increase to a score of 4 or more. The MADRS has a 10-item checklist with items rated on a scale of 0-6, for a total score range of 0 (low severity of depressive symptoms) to 60 (high severity of depressive symptoms). CGI-S measures severity of illness at the time of assessment compared with start of treatment. Scores range from 1 (normal, not at all ill) to 7 (the most extremely ill). Those who did not relapse were "censored" at their last observation.

All randomized participants are included in the time to event analyses. The numbers of participants censored are 190 and 160 for OFC and Flu groups, respectively.

Those who did not relapse were "censored" at their last observation.

All randomized participants are included in the time to event analyses. The numbers of participants censored are 217 and 220 for OFC and Flu groups, respectively.

Lack of Efficacy/Worsening of depression was at the discretion of the investigator and was based on clinical observation. Suicidality is thoughts or actions of self-harm as determined by the investigator. Those who did not relapse were "censored" at their last observation.

All randomized participants are included in the time to event analyses. The numbers of participants censored are 197 and 160 for OFC and Flu groups, respectively.

A 50% or greater improvement from baseline on the Montgomery-Asberg Depression Rating Scale (MADRS) and a Clinical Global Impressions-Severity (CGI-S) of Depression score ≤3 will be considered as response criteria met. The MADRS is a rating scale for severity of depressive mood symptoms. The MADRS has a 10-item checklist with items rated on a scale of 0-6, for a total score range of 0 (low severity of depressive symptoms) to 60 (high severity of depressive symptoms). CGI-S measures severity of illness at the time of assessment compared with start of treatment. Scores range from 1 (normal, not at all ill) to 7 (among the most extremely ill participants).

All participants who entered open-label acute treatment phase (SPII).

A 50% or greater improvement from baseline on the Montgomery-Asberg Depression Rating Scale (MADRS) and a Clinical Global Impressions-Severity (CGI-S) of Depression score ≤3 will be considered as response criteria met. The MADRS is a rating scale for severity of depressive mood symptoms. The MADRS has a 10-item checklist with items rated on a scale of 0-6, for a total score range of 0 (low severity of depressive symptoms) to 60 (high severity of depressive symptoms). CGI-S measures severity of illness at the time of assessment compared with start of treatment. Scores range from 1 (normal, not at all ill) to 7 (among the most extremely ill participants).

All participants who entered open-label stabilization treatment phase (SPIII).

Remission is defined as the Montgomery-Asberg Depression Rating Scale (MADRS) score ≤8. The MADRS is a rating scale for severity of depressive mood symptoms. The MADRS has a 10-item checklist with items rated on a scale of 0-6, for a total score range of 0 (low severity of depressive symptoms) to 60 (high severity of depressive symptoms).

All participants who entered open-label stabilization treatment phase (SPIII).

Remission is defined as the Montgomery-Asberg Depression Rating Scale (MADRS) score ≤8. The MADRS is a rating scale for severity of depressive mood symptoms. The MADRS has a 10-item checklist with items rated on a scale of 0-6, for a total score range of 0 (low severity of depressive symptoms) to 60 (high severity of depressive symptoms).

All randomized participants.

The MADRS has a 10-item checklist with items rated on a scale of 0-6, for a total score range of 0 (low severity of depressive symptoms) to 60 (high severity of depressive symptoms). Least Squares (LS) Mean values were controlled for baseline (Week 20), treatment, country, visit, and treatment by visit interaction.

All randomized participants who had baseline (Week 20) and at least 1 post-baseline (Weeks 21-47) MADRS measurements.

The MADRS has a 10-item checklist with items rated on a scale of 0-6, for a total score range of 0 (low severity of depressive symptoms) to 60 (high severity of depressive symptoms). Least Squares (LS) Mean values were controlled for baseline (Week 20), treatment and country.

All randomized participants who had baseline (Week 20) and at least 1 post-baseline (Weeks 21-47) MADRS measurements. LOCF principle was used.

CGI-S measures severity of illness at the time of assessment compared with start of treatment. Scores range from 1 (normal, not at all ill) to 7 (among the most extremely ill participants). Least Squares (LS) Mean values were controlled for baseline (Week 20), treatment, country, visit, and treatment by visit interaction.

All randomized participants who had baseline (Week 20) and at least 1 post-baseline (Weeks 21-47) CGI-S measurements.

All randomized participants who worked for pay at Week 47.

Resource utilization is defined as the average number of psychiatric visits and number of emergency room or equivalent facility visits for psychiatric illness.

All randomized participants who provided information of psychiatric visits and emergency room or equivalent facility visits for psychiatric illness from Week 21 to Week 47.

The SDS is completed by the participant and is used to assess the effect of the participant's symptoms on their work or school (Item 1), social (Item 2), and family life and home responsibilities (Item 3). Each item is measured on a 0 (not at all) to 10 (extremely) point scale with higher values indicating greater disruption. Total scores is the sum of the 3 items and range from 0 to 30 with higher values indicating greater disruption in the participant's work/social/family life. Least Squares (LS) Mean values were controlled for baseline (Week 20), treatment and country.

All randomized participants who had baseline (Week 20) and at least 1 post-baseline (Weeks 21-47) SDS score measurements. Last observation carried forward (LOCF) principle was used.

Barnes Akathisia Scale (BAS) rates observable, restless movements of drug-induced akathisia as well as the subjective awareness of restlessness and any distress associated with the akathisia. It consists of 4 items. 3 items (objective akathisia, subjective awareness of restlessness and subjective distress related to restlessness) rated on a 4-point scale, with 0 being no akathisia and 3 being severe akathisia. Item 4 (global clinical assessment of Akathisia) is derived from the responses on Items 1-3 rated on a 6-point scale, with 0 being absence and 5 being extreme Akathisia. Treatment emergent akathisia is defined as a global clinical assessment score on BAS <2 at baseline (Week 20) and a global clinical assessment score on BAS ≥2 post-baseline (Weeks 21-47).

All randomized participants who had baseline (Week 20) and at least 1 post-baseline (Weeks 21-47) BAS measurements.

Simpson-Angus Scale is used to measure Parkinsonian-type symptoms in participants exposed to neuroleptics. The scale consists of 10 items, each rated on a 5-point scale, with 0 meaning complete absence of the condition and 4 meaning the presence of the condition in extreme form. The total score is obtained by adding the items and ranges from 0-40 with higher scores indicating worse conditions. Treatment emergent parkinsonism is defined as total score ≤3 of items 1 through 10 of the Simpson-Angus scale at baseline (Week 20) and a total score >3 of items 1 through 10 post-baseline (Weeks 21-47).

All randomized participants who had baseline (Week 20) and at least 1 post-baseline (Weeks 21-47) Simpson-Angus Scale measurements.

Abnormal Involuntary Movement Scale (AIMS) is a 12-item scale designed to record the occurrence of dyskinetic movements. Items 1 through 10 are rated on a 5-point scale, with 0 being no dyskinetic movements and 4 being severe dyskinetic movements. Items 11 and 12 are yes/no questions regarding the dental condition of the participants. Treatment emergent dyskinesia is defined as a score ≥3 on any one of the AIMS items 1-7 post-baseline (Weeks 21-47) or scores ≥2 on any two of the AIMS items 1-7 post-baseline (Weeks 21-47) among participants without either criteria at baseline (Week 20).

All randomized participants who had baseline (Week 20) and at least 1 post-baseline (Weeks 21-47) AIMS measurements.

Mixed-effects model repeated measures (MMRM) analysis was used to calculate Least Squares (LS) Mean and standard error (SE). LS Mean values were controlled for baseline (Week 20), treatment, country, visit, and treatment by visit interaction.

All randomized participants who had baseline (Week 20) and at least 1 post-baseline (Weeks 21-47) cholesterol measurements.

Borderline to High fasting total cholesterol: ≥200 milligrams/deciliter (mg/dL) and <240 mg/dL at baseline and ≥240 mg/dL any time post baseline; Normal to Borderline fasting total cholesterol: <200 mg/dL at baseline, ≥200 mg/dL and <240 mg/dL any time post baseline; Normal to High fasting total cholesterol: <200 mg/dL at baseline and ≥240 mg/dL any time post baseline.

All randomized participants who had borderline or normal cholesterol level at baseline (Week 20) and at least 1 post-baseline (Weeks 21-47) cholesterol measurements.

Mixed-effects model repeated measures (MMRM) analysis was used to calculate Least Squares (LS) Mean and standard error (SE). LS Mean values were controlled for baseline (Week 20), treatment, country, visit, and treatment by visit interaction.

All randomized participants who had baseline (Week 20) and at least 1 post-baseline (Weeks 21-47) LDL cholesterol measurements.

Borderline to High fasting LDL cholesterol: ≥100 milligrams/deciliter (mg/dL) and <160 mg/dL at baseline and ≥160 mg/dL any time post baseline; Normal to Borderline fasting LDL cholesterol: <100 mg/dL at baseline, ≥100 mg/dL and <160 mg/dL any time post baseline; Normal to High fasting LDL cholesterol: <100 mg/dL at baseline and ≥160 mg/dL any time post baseline.

All randomized participants who had borderline or normal LDL cholesterol value at baseline (Week 20) and at least 1 post-baseline (Weeks 21-47) LDL cholesterol measurements.

Mixed-effects model repeated measures (MMRM) analysis was used to calculate Least Squares (LS) Mean and standard error (SE). LS Mean values were controlled for baseline (Week 20), treatment, country, visit, and treatment by visit interaction.

All randomized participants who had baseline (Week 20) and at least 1 post-baseline (Weeks 21-47) HDL cholesterol measurements.

Normal to Low fasting HDL cholesterol is ≥40 milligrams/deciliter (mg/dL) at baseline and <40 mg/dL anytime post baseline.

All randomized participants who had normal HDL cholesterol value at baseline (Week 20) and at least 1 post-baseline (Weeks 21-47) HDL cholesterol measurements.

Participants with alanine aminotransferase (ALT) and aspartate aminotransferase (AST) <=3 times the upper limit of normal (ULN) at baseline, with ALT or AST >=3 times the ULN post-baseline and total bilirubin >=2 times ULN at the same time are considered having treatment-emergent hepatic events.

All randomized participants who had baseline (Week 20) and post-baseline (Weeks 21-47) hepatic function measurements.

Mixed-effects model repeated measures (MMRM) analysis was used to calculate Least Squares (LS) Mean and standard error (SE). LS Mean values were controlled for baseline (Week 20), treatment, country, visit, and treatment by visit interaction.

All randomized participants who had baseline (Week 20) and at least 1 post-baseline (Weeks 21-47) triglycerides measurement.

Borderline to High fasting triglycerides: ≥150 milligrams/deciliter (mg/dL) and <200 mg/dL at baseline and ≥200 mg/dL any time post baseline; Normal to Borderline fasting triglycerides: <150 mg/dL at baseline, ≥150 mg/dL and <200 mg/dL any time post baseline; Normal to High fasting triglycerides: <150 mg/dL at baseline and ≥200 mg/dL any time post baseline.

All randomized participants who had borderline or normal triglycerides value at baseline (Week 20) and at least 1 post-baseline (Weeks 21-47) triglyceride measurements.

Mixed-effects model repeated measures (MMRM) analysis was used to calculate Least Squares (LS) Mean and standard error (SE). LS Mean values were controlled for baseline (Week 20), treatment, country, visit, and treatment by visit interaction.

All randomized participants who had baseline (Week 20) and at least 1 post-baseline (Weeks 21-47) glucose measurements.

Impaired to High fasting glucose: ≥100 milligrams/deciliter (mg/dL) and <126 mg/dL at baseline and ≥126 mg/dL any time post baseline; Normal to High glucose: <100 mg/dL at baseline and ≥126 mg/dL any time post baseline; Normal to Impaired fasting glucose is <100 mg/dL at baseline, ≥100 mg/dL and <126 mg/dL any time post baseline; Normal/Impaired to High fasting glucose: <126 mg/dL at baseline and ≥126 mg/dL any time post baseline.

All randomized participants who had impaired or normal glucose value at baseline (Week 20) and at least 1 post-baseline (Weeks 21-47) glucose measurements.

Mixed-effects model repeated measures (MMRM) analysis was used to calculate Least Squares (LS) Mean and standard error (SE). LS Mean values were controlled for baseline (Week 20), treatment, country, visit, and treatment by visit interaction.

All randomized participants who had baseline (Week 20) and at least 1 post-baseline (Weeks 21-47) weight measurements.

All randomized participants who had Week 20 and at least 1 post-baseline (Weeks 21-47) weight measurements.

Columbia Suicide Rating Scale (C-SSRS) captures occurrence, severity, and frequency of suicide-related thoughts and behaviors. Suicidal ideation: a "yes" answer to any one of 5 suicidal ideation questions: wish to be dead, and 4 different categories of active suicidal ideation. Suicidal behavior: a "yes" answer to any of 5 suicidal behavior questions: preparatory acts or behavior, aborted attempt, interrupted attempt, actual attempt, and completed suicide.

All randomized participants who had baseline (Week 20) and at least 1 post-baseline (Weeks 21-47) C-SSRS measurements.

Least Squares (LS) Mean values were obtained from a mixed model repeated measures (MMRM) analysis. Model includes baseline (Week 20), treatment, country, visit, and treatment by visit interaction.

All randomized participants who had baseline (Week 20) and at least 1 post-baseline (Weeks 21-47) electrocardiogram (ECG) measurements.

Data presented are the percent of participants whose baseline corrected (for rate) cardiac QT interval <500 msec with post-baseline corrected (for rate) cardiac QT interval ≥500 msec.

All randomized participants who had <500 msec QTc interval at baseline (Week 20) and at least 1 post-baseline (Weeks 21-47) electrocardiogram (ECG) measurements.

All randomized participants who had baseline (Week 20) and at least 1 post-baseline (Weeks 21-47) electrocardiogram (ECG) measurements.

Relapse is defined as meeting any of the following criteria (Relapse-any reason): 50% increase in MADRS score from randomization with concomitant CGI-S of Depression score increase to a score of 4 or more (MADRS score/CGI-S Depression Score); Hospitalization for depression or suicidality; Discontinuation due to lack of efficacy/worsening of depression/suicidality. MADRS is a rating scale for severity of depressive mood symptoms with 10 items rated on a scale of 0-6, for a total score range of 0 (low severity of depressive symptoms) to 60 (high severity of depressive symptoms). CGI-S measures severity of illness at the time of assessment compared with start of treatment. Scores range from 1 (normal, not at all ill) to 7 (the most extremely ill). Lack of Efficacy/Worsening of depression was at the discretion of the investigator and based on clinical observation. Suicidality is thoughts or actions of self-harm as determined by the investigator.

All randomized participants.