The MCCB assesses 7 cognitive domains, including speed of processing, attention vigilance, working memory, verbal learning, visual learning, reasoning and problem solving, and social cognition. T-scores in the general population have a mean of 50 and standard deviation of 10, and a higher score indicates better cognition. The primary analysis was a restricted maximum likelihood (REML) based approach using a mixed-effects model for repeated measurements (MMRM), which included the fixed categorical effects of treatment at each visit, fixed categorical effect of the stratification factor using the screening MCCB overall composite T-score, and a fixed effect for the continuous covariate of baseline at each visit. Visit was treated as the repeated measure with an unstructured covariance structure used to model the within-subject dependencies. Intercurrent events were addressed using different pre-defined strategies.

Randomized Set (RS): all patients who signed informed consent and were randomized into the trial, regardless of whether they were treated with trial medication. Patients randomized in error and discontinued from the study before the start of trial medication were excluded from the RS.

SCoRS is a 20-item interview-based assessment of cognitive deficits and the degree to which they affect day-to-day functioning. Each item is rated on a 4-point scale. Higher ratings reflect a greater degree of impairment. The interviewer integrates information from separate patient and study partner interviews to generate a total score, which ranges from 20 to 80. The analysis was a REML-based approach using a MMRM model, which included the discrete fixed effects of treatment at each visit, fixed categorical covariate of the stratification factor using the screening MCCB overall composite T-score, a fixed effect for the continuous covariate of baseline at each visit. Visit was treated as the repeated measure with an unstructured covariance structure to model the within-subject measurements. Subjects were considered as a random effect. Intercurrent events were addressed using different pre-defined strategies.

Randomized Set (RS): all patients who signed informed consent and were randomized into the trial, regardless of whether they were treated with trial medication. Patients randomized in error and discontinued from the study before the start of trial medication were excluded from the RS.

The VRFCAT is a virtual reality shopping trip performed on a tablet, and was used as an electronic Functional Capacity measure by measuring the total time adjusting for the number of errors. T-scores in the general population have a mean of 50 and standard deviation of 10, and a higher score indicates a better functional outcome. The analysis was a REML-based approach using a MMRM model, which included the discrete fixed effects of treatment at each visit, fixed categorical covariate of the stratification factor using the screening MCCB overall composite T-score, a fixed effect for the continuous covariate of baseline at each visit. Visit was treated as the repeated measure with an unstructured covariance structure to model the within-subject measurements. Subjects were considered as a random effect. Intercurrent events were addressed using different pre-defined strategies.

Randomized Set (RS): all patients who signed informed consent and were randomized into the trial, regardless of whether they were treated with trial medication. Patients randomized in error and discontinued from the study before the start of trial medication were excluded from the RS.

This is an Executive Functions/Reasoning and Problem Solving test where patients were shown two images on opposite sides of a tablet screen. Each image showed a different configuration of 3 colored balls arranged on 3 pegs. Patients were required to accurately determine the total number of times the balls in one picture would have to be moved in order to make the arrangement of balls identical to that of the other opposing picture, while employing the standard rules employed in tower tests. T-scores in the general population have a mean of 50 and standard deviation of 10, and a higher score indicates a better outcome. The analysis was performed with an analysis of covariance (ANCOVA) model, which included treatment, stratification factor of screening MCCB overall composite T-score (< 30, ≥ 30), and baseline number of correct responses on Tower of London T-score.

Randomized Set (RS): all patients who signed informed consent and were randomized into the trial, regardless of whether they were treated with trial medication. Patients randomized in error and discontinued from the study before the start of trial medication were excluded from the RS. Only patients with available data at Week 26 were included in the analysis.

PRECIS consists of 26 items covering 6 domains (memory, communication, self-control, executive function, attention, and sharp thinking), and 2 additional items assessing the overall degree of bother associated with all domains. Questions are answered via a 5-category Likert scale, with higher scores indicating a worse patient experience. The total score, ranging from 26 to 130, is the average score of the first 26 items. The analysis was a REML-based approach using a MMRM model, which included the discrete fixed effects of treatment at each visit, fixed categorical covariate of the stratification factor using the screening MCCB overall composite T-score, and continuous fixed effects for the corresponding baseline endpoint value at each visit. Visit was treated as the repeated measure with an unstructured covariance structure to model the within-subject measurements. Subjects were considered as a random effect. Intercurrent events were addressed using different pre-defined strategies.

Randomized Set (RS): all patients who signed informed consent and were randomized into the trial, regardless of whether they were treated with trial medication. Patients randomized in error and discontinued from the study before the start of trial medication were excluded from the RS.

The Humphrey visual field is a diagnostic test to measure visual fields, or perimetry. The Humphrey visual field test measures the entire area of peripheral vision that can be seen while the eye is focused on a central point. During this test, lights of varying intensities appear in different parts of the visual field while the patient's eye is focused on a central spot. The perception of these lights is charted and then compared to results of a healthy eye at the same age of the patient to determine if any damage has occurred. Visual field Index goes from 100%= perfect to 0= no vision.

Ocular sub-study Set (EYE): all treated patients who consented to participate in the ocular sub-study (including late/retrospective consent to the ocular sub-study) and had evaluable ophthalmologic measurements.

The central retinal thickness measurements were recorded for each eye via high definition optical coherence tomography (spectral domain OCT) to evaluate the retinal and sub-retinal structures.

Ocular sub-study Set (EYE): all treated patients who consented to participate in the ocular sub-study (including late/retrospective consent to the ocular sub-study) and had evaluable ophthalmologic measurements.