Background: Patients with asthma are heterogeneous in clinical presentation and in response to treatment. Despite this, tools to guide treatment are limited and include mainly measures of eosinophilic inflammation and symptoms. Airway hyperresponsiveness (AHR) to mannitol is present in patients across inflammatory phenotypes and improve with inhaled corticosteroids. Objective: To investigate whether measuring AHR to mannitol in addition to eosinophilic inflammation and symptoms adds information to the phenotypic characterization of patients with asthma. Methods: A total of 317 patients with asthma from 6 different cohorts were included in the analysis. All patients had measures of AHR to mannitol, blood eosinophils, and Asthma Control Questionnaire 5 available. A cluster analysis using Ward minimum variance method was performed. The distribution of fraction of exhaled nitric oxide, immunoglobulin E, lung function, induced sputum inflammatory cell count, age of onset, and severity of disease was compared between clusters. Results: Four clusters were identified. Three of the clusters had proportionate levels of AHR, eosinophilic inflammation, and symptoms, but 1 cluster presented with low levels of eosinophilic inflammation and a significant symptom burden. Half of the subjects in this cluster presented with AHR to inhaled mannitol. Lung function, fraction of exhaled nitric oxide, body mass index, and immunoglobulin E were normal. Conclusions: Information on AHR to mannitol in addition to blood eosinophils and symptoms identifies a subgroup of asthma patients with symptomatic, noneosinophilic disease. Airway hyperresponsiveness to mannitol may provide a treatable trait in a subgroup of patients with noneosinophilic asthma.