Multidimensional characterization of allergic rhinitis in Mysuru, South India: A cluster-based approach

Background Allergic rhinitis (AR) affects approximately 10% of Indian adults, but its clinical and biological heterogeneity remains poorly defined. Objective We sought to characterize AR phenotypes in South Indian adults using clinical features, spirometry, fractional exhaled nitric oxide (Feno), skin prick tests, blood biomarkers, and cluster analysis. Methods This was a prospective observational study of 122 patients (≥18 years old) with AR with or without asthma attending a South Indian tertiary allergy clinic and 50 asymptomatic nonatopic control subjects. Participants underwent standardized symptom/exposure assessments, spirometry, Feno, blood counts, serum IgE, and skin prick tests to 10 aeroallergens. Principal component analysis, correlation networks, and unsupervised k-means clustering were applied to define phenotypes. Results Patients with AR were younger than control subjects, predominantly female, and frequently exposed to incense and mosquito repellents. Sensitization to house dust mite (>70%) and polysensitization were common. Compared with control subjects, patients with AR had elevated Feno (43 vs18 ppb; P < .01) but no significant differences in serum IgE or eosinophils. Spirometry revealed modestly lower FEV1/forced vital capacity in patients with AR (P < .01), though values remained within normal limits. Cluster analysis identified 3 subgroups: high Feno, low eosinophils, moderate-to-severe AR with less asthma; low Feno, high eosinophils, with more asthma; and impaired lung function with moderate eosinophilia and the highest asthma burden. Network analysis demonstrated strong cosensitization between house dust mite, Parthenium hysterophorus (weed pollen), Cynodon dactylon (Bermuda pollen), and cockroach. Conclusions This study showed 3 distinct AR clusters with high dust mite sensitization alongside cosensitization with cockroach, weed, and Bermuda pollens. Multicenter studies are warranted to further refine AR clusters, including tissue-level biomarker profiling and relevance to pharmacotherapy and immunomodulatory therapies.