Non-invasive prediction of dupilumab response in facial atopic dermatitis using early post-treatment reflectance confocal microscopy features.

BACKGROUNDNon-invasive prediction of the efficacy of dupilumab on facial lesions in patients with atopic dermatitis (AD) was unresolved.

OBJECTIVETo explore reflectance confocal microscopy (RCM), a non-invasive method, for predicting dupilumab efficacy in treating facial AD lesions and to achieve effective prediction via deep learning.

METHODSFifty-two AD patients received dupilumab treatment between May 2023 and June 2025 were enrolled. Patients were categorized into 'responder' and 'non-responder' groups based on whether they achieved 75% improvement from baseline in Eczema Area and Severity Index at week 16. RCM was utilized to evaluate facial lesions. Deep learning was employed to construct a prediction model for identifying intergroup differences.

RESULTSDermal papillary dilation (area under the curve = 0.83, P < .001) and tortuous papillary capillary dilation (area under the curve = 0.81, P < .001) at week 4 correlated significantly with therapeutic non-response. The ResNet101-based deep learning model, trained on patients' RCM images of week 4 post-treatment scans, had a final test area under the curve of 0.796, with heatmaps illustrating its decision-making basis.

LIMITATIONSThe cohort size is relatively small.

CONCLUSIONEarly post-treatment RCM features can effectively predict the efficacy of dupilumab in facial AD lesions combined with deep learning.

Abstract from PubMed