BACKGROUND: Ingenol mebutate (IngMeb) 0.015% gel is an approved field treatment option for non-hyperkeratotic non-hypertrophic actinic keratosis (AK) of face and scalp. Efficacy of IngMeb has been assessed only on a clinical ground, in the majority of studies. Dermoscopy is a pivotal tool for the diagnosis of AK, while its role in evaluating the response to non-surgical therapies for AK has not been fully defined.OBJECTIVES: Our study aims to determine whether some dermoscopic features of AK of the face and scalp areas may independently predict the response to IngMeb therapy.METHODS: Clinical and dermoscopic responses, 1 month after 0.015% IngMeb therapy, were retrospectively evaluated using a per-patient and per-lesion approach. Safety was evaluated through local skin reaction composite score calculation. Demographic, clinical and dermoscopic factors were then evaluated via univariate and multivariate logistic regression analysis to assess independent predictors of response.RESULTS: Fifty-five patients with 245 AKs were enrolled. Clinically, per-patient response evaluation identified 25 (45.4%) poor/partial and 30 (54.5%) complete responders, corresponding on a per-lesion approach to 66 (26.9%) and 179 (73.1%) AKs, respectively. Dermoscopy reclassified 14 patients in the per-patient and 48 AKs in the per-lesion analysis from complete to poor/partial responders. Multivariate logistic regression analysis showed that AKs dermoscopically characterized by red pseudonetwork and located on the face were independently associated with a complete dermoscopic response to 0.015% IngMeb therapy, while microerosions were negative predictors.CONCLUSION: Specific dermoscopic features of AK may predict the response to 0.015% IngMeb therapy, together with the location on the face.