1. Removed the following secondary efficacy endpoints: change from baseline to week 12 in Zarit Burden Interview (ZBI), Cornell Scale for Depression in Dementia (CSDD), and General Medical Health Rating (GMHR). 2. Removed the Time Up and Go (TUG) test from the safety assessments. 3. Clarified that the CSDD and TUG test administered only at the Screening visit and not at follow-up visits. 4. The CSDD used as a screening tool for major depression and the TUG test as a tool for assessing risk of falls. 5. Removed the ADAS-cog, MMSE, ADCS-CGIC Overall, CSDD, and TUG test assessment at Visit 4 (Week 6). 6. Also, removed the ADCS-CGIC Overall assessment at Visit 4 (Week 6) and clarified that change from baseline will be assessed only at Visit 6 (Week 12).

1. Provided a single, global protocol for use in all countries. 2. The number of study centers was revised from 70 centers outside the US to 90 centers worldwide (including North America). 3. Alzheimer’s Disease Assessment Scale – Cognitive Subscale (ADAS-cog) removed as secondary efficacy measure. 4. Removed reference to Study 12-AVR-131. 5. Inclusion criteria updated to include clarification that a subject’s authorized representative was based on the applicable local regulations.

1. The study objectives were clarified with descriptions of primary and secondary objectives. 2. Added clarification defining the end of trial as the date of the last patient last visit. 3. Country removed from randomization stratification due to potential randomization unbalance among treatment groups. 4. Updated to remove reference to study 15-AVP-786-303 and included a reference to a potential long-term extension study for some subjects. 5. Inclusion Criterion: #9 was updated to include a full definition of women of childbearing potential and clarify acceptable forms of birth control. A specific statement was also added to clarify that women who were lactating, pregnant or plan to become pregnant were excluded. 6. Exclusion Criteria: #6 and #7 were clarified as sub-bullets under Exclusion Criteria #5. In addition, the information about the review of the electrocardiogram (ECG) for screening and baseline were updated to define how the reviews will be performed to determine eligibility for study enrollment.

1.Removed secondary endpoints: Modified Alzheimer’s Disease Cooperative Study – Clinical Global Impression of Change Rating (mADCS-CGIC-Agitation), Alzheimer’s Disease Cooperative Study – Clinical Global Impression of Change Rating (ADCS-CGIC Overall), Dementia Quality of Life (DEMQOL), Resource Utilization in Dementia (RUD). 2. Sample size calculations have been updated to address an increase in the dropout and non-evaluable rate from 15% to 20%. 3. Removed the mADCS-CGIC-Agitation, DEMQOL,RUD, and ADCS-CGIC-Overall assessments. 4. Removed ‘in clinic’ requirement for administration of study medication at visits other than Baseline. 5. Clarified terminology for the development of AVP-786 for the treatment of neuropsychiatric conditions. 6. Updated the number of subjects enrolled to 550. 7. Randomization/Stratification updated to reduce and optimize the type and number of stratification factors. 8. Inclusion Criteria and exclusion criteria were updated. 9. Removed information indicating that concomitant use of short acting benzodiazepines was allowed, as these were no longer permitted. 10. Removed the allowance to use rescue medication (lorazepam).

1. Sponsor information was updated from Avanir Pharmaceuticals, Inc. to Otsuka Pharmaceutical Development & Commercialization, Inc. 2. Summary of Changes table was added.

1. Brexpiprazole was added to the list of Prohibited Concomitant Medications. 2. The definition of the mITT population was updated for consistency with the statistical analysis plan (SAP).