Severe asthmatic patients are characterized by an increase in both exacerbation frequency and bronchial smooth muscle (BSM) mass. Rhinovirus (RV) infection of the bronchial epithelium (BE) is the main trigger of asthma exacerbations. A decreased distance between BE and BSM mass is associated with the severity of asthma. We hypothesized that RV infection of the BE increased the migration of asthmatic BSM cells, specifically. Serums, biopsies or BSM cells were obtained from 86 severe asthmatic patients and 31 non-asthmatic subjects. BE cells from non-asthmatic subjects were cultured in air-liquid interface and exposed to RV-16. Migration of BSM cells was assessed in response to BE supernatant using chemotaxis assays and live microscopy. Chemokines concentrations were analyzed by transcriptomic and ELISA. Immunocytochemistry, western blot and flow cytometry were used to quantify CXCR3 isoforms distribution. CXCR3 downstream signaling pathways were assessed by calcium imaging and western blot. BSM cells from severe asthmatic patients harbor a specific migration toward RV-infected BE whereas those from non-asthmatic subjects do not. This specific migration is driven by BE CXCL10 production that is increased both in vitro in response to RV infection and in vivo in serum from exacerbating severe asthmatic patients. The mechanism is related to both decreased expression and activation of CXCR3-B specific isoform in BSM cells from severe asthmatic patients. We demonstrate a new mechanism of BSM remodeling in severe asthmatic patients following RV exacerbation. This study highlights the CXCL10/CXCR3-B axis as a new potential therapeutic target in severe asthma.