Résumé de communication

Objectives: We aimed to systematically examine line-field confocal optical coherence tomography (LC-OCT) in keratinocyte carcinomas (KCs). The objectives were to i) unify terminology for KC image features and align it with established histopathology criteria, ii) assess interobserver agreement (IOA), iii) determine utility for triaging equivocal skin lesions, and iv) examine the associations between LC-OCT features and BCC-specific gene expression. Introduction: KCs, i.e., basal cell carcinoma (BCC) and squamous cell carcinoma (SCC), are the most common cancers worldwide, and their incidence continues to increase. To address this burden, new strategies that supplement current diagnostic methods are urgently needed. Noninvasive imaging, including the use of LC-OCT for KCs, has the potential to reduce the number of biopsies needed and shorten diagnostic delays. Materials / method: A systematic review and expert consensus harmonized LC-OCT terminology and integrated image features with established histopathological criteria for KCs. Six evaluators assessed features in LC-OCT images of KCs, and IOA was measured using Conger’s kappa. Image features’ diagnostic potential was tested for equivocal skin lesions among high-risk patients (i.e., organ transplant recipients) using exhaustive Chi-squared Automatic Interaction Detection. LC-OCT features in BCC were coupled to tape-strip transcriptomics (mRNA), and differential expression analysis was performed. Results: The initial studies identified ten key LC-OCT features that correspond to established histopathologic criteria. IOA varied across features, with high agreement for lobules and severe dysplasia, and lower agreement for dermo-epidermal junction integrity. When used clinically, a feature-based triage model demonstrated high sensitivity for detecting KCs, reaching 94.4% for BCC and 87.5% for SCC among equivocal lesions. In the final exploratory analysis, two LC-OCT features—millefeuille pattern and collagen alterations—were significantly associated with 27 differentially expressed genes in BCC. Conclusion: Our findings demonstrate that key LC-OCT image features for KCs can be standardized, reliably assessed, adopted for high-risk lesion triage, and coupled with mRNA expression to enhance understanding of underlying BCC biology, underscoring LC-OCT’s potential to shape the future of dermato-oncology.