• Amend the Principal Investigator. • Inclusion /exclusion criteria amendments based on the Sponsor’s experience in Screening during the trial • Amend the requirement for sentinel patients in Cohorts 1 and 2 to reside in the CRU from Day -1 to Day 9. • Allow an alternative emollient and shower cream to be used for the washout period and throughout the study at the Investigator’s discretion. • Add a body weight measurement on Day 1 for all cohorts in Part A and Part B to allow individual calculation of dose. • Remove 12-lead electrocardiogram (ECG) and vital signs assessment at 24 hours postdose on Day 1, Day 2 (and Day 7 [Cohorts 1 and 2] or Day 14 [Cohorts 3 and 4] for 12-lead ECGs) for all cohorts in Part A, due to predose assessments being performed the following day at the same time. • Remove the 5% time window for all procedures performed at 1 and 2 hours postdose except pharmacokinetic sampling, and replace this with a ±10 minute window, for all cohorts in Part A and Part B for logistical reasons. It was also clarified that there will be no time window for predose assessments. • Clarify that four blood samples will be taken for pharmacogenomic analysis.

Incorporate updates from Version 2.0 (dated 14 September 2020) to Version 3.0 (dated 02 December 2020) as part of the overall Protocol amendment. The rationale for this substantial amendment was to address the comments from the MHRA on Version 3.0 with respect to Exclusion Criteria #14 and #20. In addition to the amendments in V3.0, the following points were amended in V3.1 of the Protocol: • Clarify the eligibility criteria in exclusion criteria points (14) and (20) to allow the Investigator to determine patient eligibility without prior consultation with the Medical Monitor.

The purpose of this substantial amendment was to amend the range of BSA of treatable skin (not including face, scalp, genital area, palms of hands or soles of feet) affected by mild to moderate AD from ≥5% to ≤15% to ≥5% to ≤30% for Cohorts 1 and 2 in Part A (inclusion criterion [9]). The rationale for this change was to ensure that the patients entering Cohorts 1 and 2 had the same severity and extent of AD as all other patients in the study. This homogeneity of the study populations was important because the primary objective of Part A of this study was to assess the safety and tolerability of increasing dose levels of BEN2293. Restricting the BSA in Cohorts 1 and 2 may therefore have led to the inclusion only of patients with more mild disease, with the potential to skew the emergent safety data from these initial cohorts. Previously, the rationale for limiting the BSA in Cohorts 1 and 2 was to facilitate the collection of efficacy data. However, the Sponsor concluded that collection of robust efficacy data from Cohorts 1 and 2 is unlikely due to the short duration of dosing (once daily for 7 days) and the limited BSA affected by AD that was to be dosed (5%). The IMP application in Cohorts 1 and 2 in Part A remained as 5% BSA healthy skin and 5% BSA AD skin. Study emollient continued to be applied to any untreated areas for the duration of the study. The Investigator maintained close monitoring of the patients for any exacerbations in their AD and managed these as described in the Protocol.

The purpose of this substantial amendment was to amend the eligibility criteria to allow patients who received a COVID-19 or other routine vaccination within 28 days of Day 1, and throughout the study, to take part in the study (exclusion criterion point [20]). Additionally, requirements for study emollient use during the washout, treatment and Follow-up periods were clarified. A time window of 2 hours postdose (± 60 minutes) was also added for skin biopsies to be taken on the last day of dosing.

The purpose of this substantial amendment was to amend the exclusion criteria, the criteria for stopping dose escalation, individual stopping criteria, baseline and on study clinical laboratory evaluations, and the sentinel observation period, in response to increased liver transaminases observed in one patient in Cohort 3. Overall, the following changes to the Protocol were made: • For Cohort 4, the sentinel patients were dosed for 14 days, and their safety data reviewed before the remainder of the cohort could be dosed. • An extra inpatient visit was added for Cohort 4 on Day 9 for additional safety assessments; this was previously just a telephone contact. • Dose escalation stopping criteria were updated to state that if one or more patients experienced an increase in ALT and/or AST to ≥2 x ULN (confirmed upon repeat) or ≥3 x ULN (no repeat required), then dosing of all enrolled patients would stop and no further patients would be enrolled. Dosing could then only recommence following approval of a substantial Protocol amendment. • The exclusion criteria were amended to exclude any patient who had AST, ALT, GGT or total bilirubin levels above the ULN at Screening or Day -1, with no repeat testing permitted. • Added additional clinical chemistry assessments for IgG and IgE at baseline for Cohort 4 onwards, and added an extra clinical laboratory evaluation sample on Day 9 for Cohort 4.

The purpose of this amendment was to amend the inclusion criteria related to %BSA affected by AD, to amend the number of sites to be used in Part B and to amend or clarify skin, urine and plasma sampling in Part B. Overall, the following changes to the Protocol were made: • The exploratory objectives and endpoints were amended to include the evaluation of BEN2293 and metabolite BEN6403 in urine in Part A only, and the evaluation of biomarkers in skin biopsies in Part B only. • Part B would be conducted across multiple sites in the United Kingdom, Europe and Israel. • The inclusion criteria were amended to include patients with AD affecting between ≥1% to ≤30% BSA of treatable skin; this was previously ≥5% to ≤30% BSA of treatable skin. • Patients in Part B were to be supplied with a double ended medicine spoon (5 mL/2.5 mL) to remove their ointment from the jar prior to application. • An externally provided IRT system would be used to randomise patients in Part B. • Urine PK sampling was removed from the Part B Schedule of Assessments. • The skin biopsy to be conducted at the Follow-up visit was removed from the Part B Schedule of Assessments. • The Part B Sampling Summary was amended to include two additional blood samples (2 x 40 mL) for MetaHeps testing, to be conducted if the hepatic transaminase threshold criteria were reached, or if there were any liver enzyme changes or trends of concern.

The purpose of this amendment was to amend the design of Part B following completion of Part A. The main changes are listed below: • The adaptive study design was updated to clarify that up to 90 patients would be enrolled to be treated in two treatment arms of approximately 45 patients each. Patients would be randomised to receive either BEN2293 ointment or matching placebo ointment with a 1:1 parallel design. This was decided based on updated sample size determination calculations following completion of Part A of the study. • A single-blind placebo run-in period was added in which patients applied placebo ointment twice daily from Day -3 to Day -1. • Patients showing an improvement of ≥3 points in mean pruritus NRS scores (worst itch over the last 24 hours) between the end of the washout phase (mean of scores on Day 5, Day -4 and Day -3 [predose]), and the run-in phase (mean of scores on Day -2, Day 1 and Day 1 [predose]) were to be excluded from starting the double blind treatment period on Day 1. • The inclusion criteria on BP was amended • Ointment would be applied to all treatable lesioned skin (up to a maximum of 30% BSA at Day 1). Study ointment would also be applied to new treatable AD lesions that arose during the study (up to a total maximum of 33% BSA), following discussion with, or assessment by, the Investigator. • The Day 35 Follow-up visit was removed as it was not required based on Part A exposure data. • There would be only one application of BEN2293 or placebo on the final day of dosing (Day 28). • Pruritus NRS (current itch) assessments were added at predose and at 2, 4 and 8 hours postdose on Day 28. • Any participant with an affected BSA requiring treatment above 33% BSA was added to the withdrawal criteria.

The main changes to the protocol are listed below: • Include the option to extend the run-in phase in Part B of the study by a maximum of 1 day, at the Investigator’s discretion. Inclusion Criteria 10 and Exclusion Criteria 32 were updated to reflect this. • Clarify that it was imperative that patients should not use the placebo ointment administered during the run-in phase before attending the clinic on the morning of Day 1 for first dosing. • Clarify that in Part B, patients would be monitored for 30 minutes in the unit after their first doses in the run-in phase and on Day 1. • Update Exclusion Criteria 21 to allow patients to take paracetamol (intermittent use of less than or equal to 2 g/day) for up to 24 hours before dosing on Day 1 of the treatment period. • Remove grapefruit juice as a potent CYP3A4 inhibitor. However, a dietary restriction for grapefruit juice remained. • Amend the Protocol to state that in countries where there was no validated translated PROMIS scale available in the local language, the assessment would not be performed. • Clarify that after Day 1, the %BSA to be treated (sum of Day 1 %BSA treated and new AD lesions in treatable areas) should be any treated, new and resolved lesions which must not have exceeded 33% BSA after Day 1 postdose, and that this would be documented. • Clarify that patients in Part B must have refrained from consuming poppy seeds 48 hours prior to Screening and Day -3 to avoid a positive result on the drugs of abuse screen.