Identically looking placebo capsules bid (after a four-week regimen designed to mimic standard dose titration protocol)

This is a measure of rote short-term verbal memory. Total number of items correct were used as the dependent variable. The values for this measure have been recorded as change in score from baseline (T1) to after the treatment (T2). The minimum value for this scale is 0 and the maximum value is 38; higher scores mean a better outcome.

This is a measure of non-verbal memory. Total number correct across the two series of items presented was used as the dependent variable. We used the PRM total scale in this study, which represents the sum of the PRM correct scores (ranging from 0 to 24) and the PRM delayed scores (ranging from 0 to 24). Therefore, the range of the PRM total scale is from 0 to 48; higher values mean better outcomes.

The test requires participants to search under a series of colored boxes to locate a "blue token" hidden underneath one of them. During a series of trials, the participant is told that the token will be in a new location each time and that they should not go back to a location he or she has looked in previously. The main dependent variable was the total number of errors ("between errors"), which indexes the number of times a participant went back to a box where a token had already been found, lower scores mean better performance. The minimum value of the Spatial Working Memory scale is 0 and the maximum value is 137 (which was computed as the equivalent to -4 standard deviations from the mean of this measure); higher scores mean worse outcomes. The values for this measure have been recorded as change in score from baseline (T1) to after the treatment (T2).

This measure is a computerized version of the Corsi Blocks task, a long-standing neuropsychological test. The main dependent variables selected for this test was the span length, which is the longest sequence of numbers recalled accurately. The minimum value of the Spatial Span Length Score Scale is 0 and the maximum value is 9; higher scores mean better outcomes. The values for this measure have been recorded as change in score from baseline (T1) to after the treatment (T2).

This is a measure of inhibitory control, often used as a marker for prefrontal-striatal function integrity. Specifically, it measures the participant's ability to inhibit pre-potent behavioral responses that have been established by provision of prior "go" or "no-go" cues in a classical conditioning paradigm. The main dependent variables selected was speed of response of execution to Go targets. The minimum value of the speed of response of execution to Go targets is 280 milliseconds (ms) and the maximum value is 1000 ms; higher scores mean worse outcomes. The values for this measure have been recorded as change in score from baseline (T1) to after the treatment (T2).

Incidence of adverse events was monitored by clinical history, physical examinations, electrocardiograms (ECGs), clinical laboratory tests, the Screen for Childhood Anxiety Related Emotional Disorders (SCARED). Here, we report the analysis of the effect of memantine treatment on QTc intervals because of its clinical importance for this analysis for potential drug toxicity. QTc intervals ≥ 450 ms are generally considered long, and drug-induced QTc interval prolongations ≥ 60 ms are generally considered clinically relevant.

This is a measure of receptive syntax skills (Bishop, 1983). Participants are asked to point to a picture (out of 4) that corresponds to a phrase or sentence spoken by the examiner. The total number of items correct (rather than blocks passed) will be used as the dependent variable, following the administration manual's ceiling rule. The values for this measure have been recorded as change in score from baseline (T1) to after the treatment (T2). The minimum value of the scores is 0 and the maximum value is 40; with higher scores considered to be a better outcome.

This is a measure of receptive semantics, whereby the participant is asked to point to a picture (out of 4) that corresponds to a word spoken by the examiner. As this test has a 0.85 correlation with composite measures of Verbal IQ (i.e. from the Wechsler Intelligence Scale series), it can be used in conjunction with the Matrices subtest to estimate overall intellectual functioning. The total number of items correct was used as the dependent variable, following the administration manual's rules for basals and ceilings. The values for this measure have been recorded as change in score from baseline (T1) to after the treatment (T2). The minimum value for this scale is 0 and the maximum value is 192, higher scores mean a better outcome.

This is a measure of adaptive functioning that integrates information from 13 different domains (e.g., gross motor, social interaction, eating, toileting, dressing, personal self-care, etc.). It is in a questionnaire format, which a caregiver can complete while the participant is being tested. Standard scores for all indices will be derived from age norms that extend from birth to age 80, as these were used as dependent variables. We report here on the Broad Independence Score recorded as change in score from baseline (T1) to after the treatment (T2). The minimum value of the SIB-R Score Scale in this study was -24 (this number is below 0 because -24 was the minimum value for the worst performing participant in the trial) and the maximum value of this scale is 153; higher scores mean better outcomes.

Total of all reporting groups

Race and Ethnicity were not collected from any participant.

The primary efficacy measure is focused on episodic memory. The CVLT-II short form assesses supraspan word learning ability as an index of episodic verbal long-term memory. We hypothesize that treatment with memantine will produce significant improvements in this test. The main dependent variable selected, based on prior literature was the total number of target items correct summed across learning trials 1-4. Scale Range: from 0 to 36; higher scores represent higher cognitive functioning.

This test provides a measure of non-verbal reasoning ability that requires subjects to visually inspect a matrix of 4 or 9 pictures that has a missing piece. Participants infer a rule or pattern in the stimuli and select the appropriate response from a range of 4-6 possibilities. Since age norms are not available for individuals older than 17y11m, the ability score was used. This score based on Rasch modeling that corrects for different items sets being administered to participants. The minimum value of the DAS-II Score is 0 and the maximum is 153. Higher values represent a better outcome.

This is a measure of adaptive functioning that integrates information from 13 different domains (e.g., gross motor, social interaction, eating, toileting, dressing, personal self-care, etc.). It is in a questionnaire format, which a caregiver can complete while the participant is being tested. The minimum value of the SIB-R Score Scale in this study was -24 (this number is below 0 because -24 was the minimum value for the worst performing participant in the trial) and the maximum value of this scale is 153; higher scores mean higher adaptive functioning.

This is a measure of receptive semantics. The participant is asked to point to a picture (out of 4) that corresponds to a word spoken by the examiner. As this test has a 0.85 correlation with composite measures of Verbal IQ (i.e. from the Wechsler Intelligence Scale series), it can be used to estimate overall intellectual functioning. The total number of items correct was used as the dependent variable, following the administration manual's rules for basals and ceilings. The minimum value for this scale is 0 and the maximum value is 192, higher scores mean higher intellectual functioning.

Memantine were dispensed in a 66-day supply (56 days plus 10 extra days) by the study coordinator to participants receiving the placebo at the end of visits 2 and 3.

Memantine: Encapsulated Memantine 10 mg bid (after four-week standard dose titration protocol)

This is a measure of non-verbal memory that requires the participant to learn associations between an abstract visual pattern and its location. Two dependent variables have been selected: Total number of items correct on the first trial of each stage, and total number of stages completed. The values for this measure have been recorded as change in score from baseline (T1) to after the treatment (T2). The minimum value of the PAL Memory Score Scale is 0 and the maximum value is 21; higher scores mean better outcomes.

The analysis population consisted of participants who completed 16 weeks of treatment and had neuropsychological assessments at baseline (T1) and after treatment (T2)

This is a measure of rote short-term verbal memory. Total number of items correct were used as the dependent variable. The values for this measure have been recorded as change in score from baseline (T1) to after the treatment (T2). The minimum value for this scale is 0 and the maximum value is 38; higher scores mean a better outcome.

The analysis population consisted of participants who completed 16 weeks of treatment and had neuropsychological assessments at baseline (T1) and after treatment (T2).

This is a measure of non-verbal memory. Total number correct across the two series of items presented was used as the dependent variable. We used the PRM total scale in this study, which represents the sum of the PRM correct scores (ranging from 0 to 24) and the PRM delayed scores (ranging from 0 to 24). Therefore, the range of the PRM total scale is from 0 to 48; higher values mean better outcomes.

The analysis population consisted of participants who completed 16 weeks of treatment and had neuropsychological assessments at baseline (T1) and after treatment (T2).

The test requires participants to search under a series of colored boxes to locate a "blue token" hidden underneath one of them. During a series of trials, the participant is told that the token will be in a new location each time and that they should not go back to a location he or she has looked in previously. The main dependent variable was the total number of errors ("between errors"), which indexes the number of times a participant went back to a box where a token had already been found, lower scores mean better performance. The minimum value of the Spatial Working Memory scale is 0 and the maximum value is 137 (which was computed as the equivalent to -4 standard deviations from the mean of this measure); higher scores mean worse outcomes. The values for this measure have been recorded as change in score from baseline (T1) to after the treatment (T2).

The analysis population consisted of participants who completed 16 weeks of treatment and had neuropsychological assessments at baseline (T1) and after treatment (T2).

This measure is a computerized version of the Corsi Blocks task, a long-standing neuropsychological test. The main dependent variables selected for this test was the span length, which is the longest sequence of numbers recalled accurately. The minimum value of the Spatial Span Length Score Scale is 0 and the maximum value is 9; higher scores mean better outcomes. The values for this measure have been recorded as change in score from baseline (T1) to after the treatment (T2).

The analysis population consisted of participants who completed 16 weeks of treatment and had neuropsychological assessments at baseline (T1) and after treatment (T2).

This is a measure of inhibitory control, often used as a marker for prefrontal-striatal function integrity. Specifically, it measures the participant's ability to inhibit pre-potent behavioral responses that have been established by provision of prior "go" or "no-go" cues in a classical conditioning paradigm. The main dependent variables selected was speed of response of execution to Go targets. The minimum value of the speed of response of execution to Go targets is 280 milliseconds (ms) and the maximum value is 1000 ms; higher scores mean worse outcomes. The values for this measure have been recorded as change in score from baseline (T1) to after the treatment (T2).

The analysis population consisted of participants who completed 16 weeks of treatment and had neuropsychological assessments at baseline (T1) and after treatment (T2). Notice that, for this specific test, some participants stopped midway.

Incidence of adverse events was monitored by clinical history, physical examinations, electrocardiograms (ECGs), clinical laboratory tests, the Screen for Childhood Anxiety Related Emotional Disorders (SCARED). Here, we report the analysis of the effect of memantine treatment on QTc intervals because of its clinical importance for this analysis for potential drug toxicity. QTc intervals ≥ 450 ms are generally considered long, and drug-induced QTc interval prolongations ≥ 60 ms are generally considered clinically relevant.

The analysis population consisted of all participants who started treatment and had good-quality ECGs before and after the start of memantine treatment, whether or not they completed the entire treatment.

This test provides a measure of non-verbal reasoning ability that requires subjects to visually inspect a matrix of 4 or 9 pictures that has a missing piece. Participants have to infer a rule or pattern in the stimuli and select the appropriate response from a range of 4-6 possibilities. Since age norms are not available for individuals older than 17y11m, the ability score will be used as the dependent variable. This is an intermediate score based on Rasch modeling that corrects for different items set being administered to participants. The minimum value of the DAS-II Rasch Score Scale is 0 and the maximum value is 153; higher scores mean better outcomes. The values for this measure have been recorded as change in score from baseline (T1) to after the treatment (T2).

The analysis population consisted of participants who completed 16 weeks of treatment and had neuropsychological assessments at baseline (T1) and after treatment (T2).

This is a measure of receptive syntax skills (Bishop, 1983). Participants are asked to point to a picture (out of 4) that corresponds to a phrase or sentence spoken by the examiner. The total number of items correct (rather than blocks passed) will be used as the dependent variable, following the administration manual's ceiling rule. The values for this measure have been recorded as change in score from baseline (T1) to after the treatment (T2). The minimum value of the scores is 0 and the maximum value is 40; with higher scores considered to be a better outcome.

The analysis population consisted of participants who completed 16 weeks of treatment and had neuropsychological assessments at baseline (T1) and after treatment (T2).

This is a measure of receptive semantics, whereby the participant is asked to point to a picture (out of 4) that corresponds to a word spoken by the examiner. As this test has a 0.85 correlation with composite measures of Verbal IQ (i.e. from the Wechsler Intelligence Scale series), it can be used in conjunction with the Matrices subtest to estimate overall intellectual functioning. The total number of items correct was used as the dependent variable, following the administration manual's rules for basals and ceilings. The values for this measure have been recorded as change in score from baseline (T1) to after the treatment (T2). The minimum value for this scale is 0 and the maximum value is 192, higher scores mean a better outcome.

The analysis population consisted of participants who completed 16 weeks of treatment and had neuropsychological assessments at baseline (T1) and after treatment (T2).

This is a measure of adaptive functioning that integrates information from 13 different domains (e.g., gross motor, social interaction, eating, toileting, dressing, personal self-care, etc.). It is in a questionnaire format, which a caregiver can complete while the participant is being tested. Standard scores for all indices will be derived from age norms that extend from birth to age 80, as these were used as dependent variables. We report here on the Broad Independence Score recorded as change in score from baseline (T1) to after the treatment (T2). The minimum value of the SIB-R Score Scale in this study was -24 (this number is below 0 because -24 was the minimum value for the worst performing participant in the trial) and the maximum value of this scale is 153; higher scores mean better outcomes.

The analysis population consisted of participants who completed 16 weeks of treatment and had neuropsychological assessments at baseline (T1) and after treatment (T2). Note that some participants chose not to undergo the questionnaire.

Memantine capsules were dispensed in a 66-day supply (56 days plus 10 extra days) by the study coordinator to participants receiving the placebo at the end of visits 2 and 3.

Memantine: Encapsulated Memantine 10 mg bid (after four-week standard dose titration protocol)

Clinically significant increase in thyroid-stimulating hormone concentrations in participant with hypothyroidism between screening and follow-up visits