Background: Type 2 inflammation is characterized by enhanced activity of interleukin (IL)-4, -5 and -13, and treatments targeting these pathways are available for treatment of severe asthma. At present, the pattern of pathway activity and the implications overlapping of pathway activity are unknown. Objective: We hypothesized that clustering of airway mRNA expression would identify distinct molecular subtypes of severe asthma and thereby uncover the prevalence and overlap of pathway activity. Methods: Sputum mRNA expression of genes related to expression of IL-5(CLC, CPA3 and DNASE1L3), IL-13(IL13Ra1, TNFSF14 and SERPINB2), T1/Th17 activity(IL1B, ALPL and CXCR2) and in vitro response to corticosteroids (FKBP512) and mepolizumab (ARAP3) was analysed in patients (n = 109) with severe asthma and healthy controls (n = 22). A cluster analysis of gene expression was performed. The response to a short course of OCS was assessed in a subset of patients (n = 29). Results: Five molecular clusters were identified. Three had abundant T2 gene expression of which two (n = 39 and n = 9) were characterized by abundant expression of both IL-13- and IL-5-related genes. The last (n = 6) had only abundant IL-5-related gene expression. These T2-high molecular clusters could not be distinguished using T2 biomarkers. T2- and Th1/Th17-related mRNA expression were co-expressed across all clusters. OCS significantly reduced T2 gene expression (CLC, IL13Ra1, SERPINB2 and ARAP3) and significantly increase expression of Th1/Th17-related genes (ALPL and CXCR2). Conclusions and clinical relevance: Clustering of airway mRNA expression identified five molecular clusters of severe asthma of which three were considered T2 high. Co-expression of IL-5- and IL-13-related genes at moderate levels was present in almost half of patients, while marked elevated expression of both was rare. In contrast to IL-5, clusters with isolated IL-13- and Th1/Th17-related gene expression were not identified.