Pathogenic and genetic landscape of Still's disease across ages, with new insights into age-related IL-18 patterns.

Still's disease (SD) is now widely recognized as a single disease spectrum encompassing both childhood- and adult-onset forms. While most comparative studies so far have focused on clinical manifestations and treatment responses, aspects related to the pathophysiology of the disease including molecular mechanisms and genetic predisposition have remained less deeply explored.In this review, we provide an integrated overview of current knowledge on the pathogenic pathways driving SD. We summarize evidence supporting their contribution across the age spectrum from children to adults, and discuss age-related trends in IL-18 and S100 proteins, recognized key-mediators of systemic inflammation.To complement the existing literature, we present novel data on IL-18 measurements collected in the context of cohort studies and clinical practice in three tertiary centers involved in SD care across the ages, showing a modest but significant age-related decline in patient IL-18 levels across sJIA and AOSD cohorts.Finally, we emphasize that both sJIA and AOSD share convergent genetic associations within the HLA class II region, although specific variants may exert different functional effects. In particular, HLA-DRB1*15 has been linked to disease susceptibility in the adult spectrum, whereas in children this specific HLA background appears to predispose to lung involvement and features of hypersensitivity.Altogether, the data presented collectively support the view that SD represents a single acquired and complex autoinflammatory disease in which both pediatric and adult forms share core pathogenic mechanisms as well as genetic associations within the HLA class II region. Nevertheless, we observed some differences between children and adults that remain difficult to interpret, as it remains unclear whether they reflect true biological variation or result from differences in study design or methodology. Further comparative, cross-cohort and longitudinal studies will be needed to resolve these uncertainties.