A Randomised, Double-Blind, Placebo-Controlled Study Evaluating the Efficacy and Safety of Orvepitant in Subjects With Major Depressive Disorder

HAM-D was use to measure the severity of depressive symptoms in participants with primary depressive illness. It was a checklist of items that were ranked on a scale of 0-4 or 0-2. The HAM-D Total Score was calculated by summing the individual response scores. The lowest possible score was 0 (absence of depression) and the highest possible score was 52 (most severe measure of depression). For the last observation carried forward analyses, the most recent post randomization total score (as opposed to individual responses) was "carried forward" and used in the calculation of the change from randomization value. If the responses to more than 1 question were missing for a participant at a particular time point, the total score was not calculated. Data was presented as percent of HAM-D responders which was defined as participants who has a >=50% reduction from randomization in HAM-D total score.

The start of the 'maintained antidepressant response' was the time at which a participant demonstrates a 50% reduction from randomization in their HAM-D total score and where this response was maintained until the end of the treatment phase (week 6). Participants who met the 50% reduction at week 6 without having met it at week 4 were considered to have reached a maintained response, and therefore were censored at week 6. Where a participant met the criteria for maintained antidepressant response, the "time (in days) to maintained antidepressant response" was calculated as: (Date of assessment at which the maintained response commences minus Date of randomization) plus 1. Where a participant did not met the criteria for maintained antidepressant response, their time to response was censored at the last on-treatment assessment they undertake, up to and including the week 6 assessment.

The BECH scale was extracted from the HAMD-17 and comprised the 6 items (sum of items 1, 2, 7, 8, 10, and 13 of the 17-item HAM-D scale): Depressed Mood, Feelings of Guilt, Work and Activities, Retardation, Anxiety Psychic and Somatic Symptoms General. The BECH Total Score was calculated by summing the individual response scores. It was a checklist of items that were ranked on a scale of 0-4 or 0-2. The lowest possible score was 0 (absence of depression) and the highest possible score was 22 (most severe measure of depression). Due to the small number of items , missing data was not imputed for the BECH Total Score. If any of the 6 items above were missing, the total score was not calculated at that visit. Value at randomization was the Baseline value. Change from Baseline in BECH Total Score was the difference between BECH Total Score at the time point being analyzed to randomization.

QIDS-SR assessed symptoms severity of Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) diagnostic criterion for major depressive disorder. It consisted of 16 separate items, defining 9 DSM-IV symptom criterion domains. A total score was obtained by summing scores on each domain. The lowest possible score was 0, which represented an absence of depression; the highest possible score was 27, which represented the most severe measure of depression. Due to the small number of items, missing data was not imputed for the QIDS-SR total score. If any of the 9 domains above were missing, the total score was not calculated at that visit. Value at randomization was the Baseline value. Change from Baseline in total score was the difference between QIDS total score at the time point being analyzed to randomization. If no post- randomization scores were available, change from Baseline was set to missing.

The anxiety score was extracted from the HAM-D-17 and comprises of items 10, 11, 12, 13 and 15 from the HAM-D scale. The anxiety score was calculated by summing the individual response scores to these questions. It was a checklist of items that were ranked on a scale of 0-4 or 0-2. The lowest possible score was 0 (absence of depression) and the highest possible score was 18 (most severe measure of depression). Due to the small number of items, missing data was not imputed for the anxiety score. If either of the anxiety items was missing, the total score was not calculated at that visit. Value at randomization was the Baseline value. Change from Baseline in anxiety score was the difference between the anxiety score at the time point being analyzed to randomization. If no post- randomization scores were available, change from Baseline was set to missing.

The CGI-I assessed scores range from 1 - Very much Improved to 7 - Very much worse, with 0 representing a participant that was not assessed. The assessed scores were dichotomized. Scores of 1 or 2 was in the first category, scoring 1. All other scores (except zero which was regarded as missing) was in the second category, scoring 0. The percentage of participants in the first category was calculated for each assessment.

The CGI-S assessed scores range from 1 - Very much Improved to 7 - Very much worse, with 0 representing a participant that was not assessed. For the CGI-S, remote, blinded MedAvante, raters assessed the participant's severity of illness considering their total clinical experience with the particular population being studied and information obtained during the Baseline HAMD interview with the participant. Value at randomization was the Baseline value. Change from Baseline in total score was the difference between CGI-S Score at the time point being analyzed to randomization. If no post- randomization scores were available, change from Baseline was set to missing.

CPFQ was a brief self-report scale which was designed to measure cognitive and executive dysfunction in mood and anxiety disorders. The CPFQ comprised of 7 questions assessing each of the most common complaints of depressed participants reporting fatigue or cognitive/executive problems. Each question was rated on a scale of 1 to 6, with 1 indicating greater than normal functioning, 2, indicating normal functioning, and with higher numbers indicating poorer functioning. Two versions of the CPFQ were utilized during the study. The Baseline CPFQ requested the participant reflect back over the past month. For the treatment period, the CPFQ requested the participant reflect back over the past week. Value at randomization was the Baseline value. Change from Baseline in Total Score was the difference between CPFQ Score at the time point being analyzed to randomization. If no post- randomization scores were available, change from Baseline was set to missing.

The MSQ was a self-rated scale designed to assess effects on sleep and effects on next day functioning. The following variables were assessed in order to determine effects on sleep: total sleep time, sleep onset latency, number of nocturnal awakenings, wake time after sleep onset and sleep quality (from poor, assigned a score of 1, to excellent, assigned a score of 10). The refreshing value of the sleep was also determined (poor assigned a score of 1, to excellent, assigned a score of 10). Value at randomization was the Baseline value. Change from Baseline was calculated for each domain separately. Change from Baseline was the difference between score at the time point being analyzed to randomization. If no post- randomization scores were available, change from Baseline was set to missing.

The MSQ was a self-rated scale designed to assess effects on sleep and effects on next day functioning. The following variables were assessed in order to determine effects on sleep: total sleep time, sleep onset latency, number of nocturnal awakenings, wake time after sleep onset and sleep quality (from poor, assigned a score of 1, to excellent, assigned a score of 10). The refreshing value of the sleep was also determined (poor assigned a score of 1, to excellent, assigned a score of 10). Value at randomization was the Baseline value. Change from Baseline was calculated for each domain separately. Change from Baseline was the difference between score at the time point being analyzed to randomization. If no post- randomization scores were available, change from Baseline was set to missing.

The MSQ was a self-rated scale designed to assess effects on sleep and effects on next day functioning. The following variables were assessed in order to determine effects on sleep: total sleep time, sleep onset latency, number of nocturnal awakenings, wake time after sleep onset and sleep quality (from poor, assigned a score of 1, to excellent, assigned a score of 10). The refreshing value of the sleep was also determined (poor assigned a score of 1, to excellent, assigned a score of 10). Value at randomization was the Baseline value. Change from Baseline was calculated for each domain separately. Change from Baseline was the difference between score at the time point being analyzed to randomization. If no post- randomization scores were available, change from Baseline was set to missing.

A HAMD remitter was defined as a participant who had a HAMD Total Score <=7. The HAMD total score was calculated for each participant at each time point. Those participants with no missing value for HAMD total score were categorized as having a HAMD total score of <=7 or >7.

Assessment of suicidality were done through use of the CSSRS for suicidal ideation and suicidal behavior. For suicidal ideation ratings were 1 to 5, where 1. Wish to be dead, 2. Non-specific active suicidal thoughts, 3. Active suicidal ideation without intent to act, 4. Active suicidal ideation with any methods (not plan) without intent to act, 5. Active suicidal ideation with specific plan and intent and for suicidal behavior ratings were 6 to 12, Where 6. Actual attempt, 7. Engaged in non-suicidal self-injurious behavior, 8. Interrupted attempt, 9. Aborted attempt, 10. Preparatory acts or behavior, 11. Suicidal behavior, 12. Completed suicide. n= number participants with at least one CSSRS assessment after the first dose of study medication (i.e. on treatment or post treatment). Only those categories from CSSRS (1-12) are presented for which symptoms were actually observed in the participants. Categories with null values for all the arms have not been presented.

The DESS scale consisted of 43 signs and symptoms, scored as 'new symptom', 'old symptom but worse', 'old symptom but improved' or 'symptom not present/old symptom but unchanged'. The total number of new signs and symptoms, old symptoms but worse, old symptoms but improved and the total number of new or old-but-worse signs and symptoms were calculated for each treatment and visit. n = number of subjects who had at least one of the 43 symptoms in the specified category. The summary for a specified category are of the number of symptoms the n subjects had in that category. Treatment period was up to Week 6.

MSFQ included five items with a score ranging from 1 to 6 (1 = greater than normal; 2 = normal; 3 = minimally diminished; 5 = markedly diminished; and 6 = totally absent). The following areas of sexual functioning were included: diminished/absent libido; arousal difficulties; orgasm difficulties/anorgasmia; erectile dysfunction (males only) and degree of sexual satisfaction. A total score was used as a global measure of sexual dysfunction. The Baseline MSFQ requested the participant reflect back over the past month. For the treatment period, the follow-up MSFQ requested the participant reflect back over the past week. Change from Baseline was the value at post-Baseline visit minus Baseline value.

MSFQ included five items with a score ranging from 1 to 6 (1 = greater than normal; 2 = normal; 3 = minimally diminished; 5 = markedly diminished; and 6 = totally absent). The following areas of sexual functioning were included: diminished/absent libido; arousal difficulties; orgasm difficulties/anorgasmia; erectile dysfunction (males only) and degree of sexual satisfaction. A total score was used as a global measure of sexual dysfunction. The Baseline MSFQ requested the participant reflect back over the past month. For the treatment period, the follow-up MSFQ requested the participant reflect back over the past week. Change from Baseline was the value at post-Baseline visit minus Baseline value.