Two identically designed multi-centred, double-blind RCTs (AD-301 and AD-302), both entirely based in the US, met the inclusion criteria for this systematic review. Both studies randomized patients with AD scored as mild-to-moderate using ISGA, in a 2:1 ratio to either crisaborole or vehicle over a treatment course of 28 days. A larger percentage of crisaborole-treated patients versus vehicle achieved the primary outcome, treatment success according to the ISGA at day 29, and this difference was statistically significant in both studies. No conclusions can be drawn regarding HRQoL, and this is an important limitation given the impact of AD on this outcome. There was some indication of a numerically higher risk of application site pain with crisaborole, although there was no increased risk of WDAE. The lack of an active comparator in the two trials was a limitation of this review. The data available from the NMA suggests no statistically significant differences for crisaborole versus TCI for the percentage of patients achieving an ISGA of clear or almost clear; however, there were several limitations with the analysis, and no data were available to compare crisaborole to TCS.