The main protocol modifications emerged after thorough review of the statistical considerations in order not to underestimate the true treatment effect in comparison to the placebo. These considerations included an increase of the sample size from 90 randomized subjects to 125 randomized subjects, and a switch in the randomization scheme from a 2:2:1 (vafidemstat 0.6 mg/day:vafidemstat 1.2 mg/day:placebo) to a 1:1:2 scheme. Specifically, the ETHERAL trial started with a randomization scheme of 2:2:1 between two doses of vafidemstat (0.6 and 1.2 mg) and placebo during the first 6 months treatment period. However, for the following reasons a randomization ratio of 1:1:2 led to a more informative study: 1. Preclinical studies suggested that both doses should have similar therapeutic effect. If the safety and efficacy profiles of the two vafidemstat arms were similar, then a comparison of the pooled vafidemstat arms vs. the placebo group could be more powerful with a 1:1:2 allocation ratio. This combined comparison analysis was finally not performed 2. The placebo group subjects were actually under the standard of care and were expected to have an improvement. A larger proportion of subjects assigned to the placebo group and with re-allocation in the extension period allowed a more efficient within subject comparison within the placebo 3. In case future investigations with vafidemstat were carried out with a Bayesian design, with room to borrow information from previous studies, having a more balanced vafidemstat to placebo assignment in the current study may be more efficient Other changes made to the protocol were: • Planned dates of the last subject first visit (from 3Q 2018 to 1Q 2019) and the last subject last visit (from 3Q 2019 to 1Q 2020) • Accountability of the AChEI (concomitant chronic medication used by all the subjects), from being tracked in the same way of the Investigational Medical Product to being tracked as the other concomitant medications

Protocol modifications have emerged from a thorough review, where the introduction of some modifications was implemented in order to clarify items that were not previously explained with sufficient detail. The following modifications were performed: • Clarification in definition of fertile female for the inclusion criterion number 12 • Clarification of exclusion criteria number 7 and 8 • Amendment of exclusion criterion number 11 to facilitate recruitment by accommodating the wash-out period to the common criteria in clinical trials • Clarification of window period between visits • Clarification of the blood extractions to be taken during the study • Addition of two Cogstate practice assessments before baseline • Clarification of the subjects re-screened • Clarification of two criteria for removing subjects from therapy or assessment • Clarification of information given to the subject about the storage and handling of Investigational Medicinal Product • Clarification of the treatment compliance • Clarification of the order of the efficacy assessments and the administration of the Investigational Medicinal Product • Clarification of when a clinical laboratory abnormality should be documented as an adverse event

The following modifications were performed: • Inclusion of an exploratory objective, and the corresponding endpoints, to document APOE genotype in the study’s participants as well as examine any associated treatment differences. Finally, none of the planned APOE genotype relationship analyses were performed • Clarification of the Extension Period as an open period because all subjects will be on active treatment and results will be unblinded after the last subject last visit in the 24 week double blind placebo-controlled Treatment Period • Inclusion of an optional blood sample test at any point throughout the study, although baseline collection is recommended, for APOE genotyping • Amendment of inclusion criterion number 10: AD treatment-naïve subjects as eligible participants in the study (only in centers located in France and UK) • Clarification of AChEI compliance assessment as only would apply to subjects with AChEI at the Screening Visit • Amendment of withdrawal criterion about changes in the concomitant medication to make it apply to both AD treatment-naïve subjects and subjects with AChEI at the Screening Visit • Amendment of the statistical issues due to the inclusion of the APOE genotyping, and AD treatment-naïve subjects as eligible participants This protocol amendment also included changes in the study administrative structure