AI-Assisted Quantitative Reflectance Confocal Microscopy Reveals Modulation of Senescence-Associated Epidermal Morphometric Features by a Topical TGF-β Mimetic, XEP™-716 Miniprotein™.

introductionEpidermal aging is characterized by progressive structural alterations, including reduced keratinocyte density, altered cellular morphology, and widened intercellular spacing (ICS). Reflectance confocal microscopy (RCM) combined with artificial intelligence (AI)-assisted image analysis enables non-invasive quantification of these changes in vivo. This secondary analysis of RCM images reported in a previously published randomized split-face clinical trial aimed to identify quantitative morphometric features associated with chronological aging and to evaluate their modulation following topical application of the transforming growth factor-beta (TGF-β) mimetic XEP™-716 Miniprotein™ (Activen SA, Lausanne. Switzerland).

methodsPreviously acquired baseline RCM images from 49 women (age 40-60 years) were analyzed for associations with chronological age. Treatment effects were assessed in 25 women from the 2.5% arm of a 56-day randomized, double-blind, split-face study comparing XEP™-716 Miniprotein™ 2.5% with vehicle. AI-assisted image analysis quantified keratinocyte density, cellular circularity, and ICS in the stratum granulosum (SG) and the stratum spinosum (SS). Cells with circularity ≤ 0.439 were classified as low-circularity cells and those ≥ 0.440 circularity as morphologically normal. Treatment effects were assessed using linear mixed-effects models.

resultsStratum granulosum total and normal keratinocyte densities were nominally significantly associated with age (p = 0.015 and p = 0.023, respectively). Compared with vehicle, XEP™-716 Miniprotein™ used at 2.5% was associated with nominally significant increases in normal keratinocyte density (p = 0.028) and the circularity of low-circularity cells (p = 0.018), and with decreases in ICS thickness (p = 0.030) and low-circularity cell density (p = 0.044) in the SG. In the SS, there was a non-significant trend towards reduced low-circularity cell density (p = 0.075).

conclusionArtificial intelligence-assisted quantitative RCM can identify age-associated epidermal morphometric features, predominantly in the SG, and detect their modulation following topical treatment. XEP™-716 Miniprotein™ at 2.5% showed consistent treatment-associated changes across several morphometric parameters with nominal statistical significance. These exploratory findings warrant prospective validation against molecular senescence markers and functional barrier measurements.