four of them) or by RCM alone (evaluation performed by six of them) 1 year before and results were compared with the previous study. Mean sensitivity, specificity and overall diagnostic accuracy (ACC) and 95% CI were calculated for the specific diagnosis of different facial lesions (Table 1). As expected, considering the evaluation of the seven experts, the combination dermoscopy/RCM had a higher mean ACC for both malignant tumour (including LM/LMM) and LM/LMM of the face (84%, CI 80-89 and 84%, CI 79-89, respectively) than the two imaging techniques alone (74%, CI 68-80 for dermoscopy vs. 80%, CI 75-85 for RCM in case of LM/LMM and 75%, CI 68-80 for dermoscopy vs. 80%, CI 74-85 for RCM in case of malignant tumour). Notably, ACC increased for each investigator. The superiority of the ACC of the combination dermoscopy/RCM on the single imaging techniques for LM/LMM was also observed when comparing the current data to the results obtained in the previous study by all the 21 investigators. Considering all evaluations, the ACC increased from dermoscopy to RCM and to dermoscopy plus RCM. According to the CI, the combination dermoscopy/RCM improved both mean sensitivity and mean specificity of RCM alone for LM/LMM, whereas it highly improved mean sensitivity of dermoscopy alone but not its mean specificity (sensitivity of 65% CI 57-73, 82% CI 76-89 and 85% CI 80-91 and specificity of 88% CI 82-95, 77% CI 68-86 and 83% CI 75-91 for dermoscopy, RCM, and dermoscopy plus RCM, respectively). As already discussed, this fact was probably related to the presence of hyper-reflective dendritic Langerhans cells under RCM mistaken for neoplastic melanocytes in the epidermis of most misdiagnosed benign lesions. 9 Concerning the facial lesions other than LM/LMM, we also observed an increase in ACC with the combination dermoscopy/RCM (Table 1). In conclusion, this study shows that the integration of dermoscopy and RCM increases the ACC of both these techniques used alone for facial tumours. More in-depth studies should be carried out, even using machine learning techniques, for optimal integration of dermoscopic data with RCM data, in order to increase ACC by reducing possible false positives induced by the RCM examination. This study received the approval from the Ethical Committee of the University Hospital of Saint-Etienne (Institutional review board number 672016/CHUSTE).