Cytokines and immunologic checkpoint molecules in predicting success of allergen immunotherapy.

Clinical Snapshot

Key Findings

• P Value: Not significant (specific values not provided)

• Effect Size: No significant correlations identified

• Primary Outcome: Prediction of allergen immunotherapy success using pre-treatment biomarker levels

• Nnt Or Sensitivity: Clustering algorithms grouped 48/60 subjects into 3 clusters with no clinical correlation

• Confidence Interval: Not reported

Clinical Application

Limited - study demonstrates that extensive cytokine and checkpoint molecule profiling does not predict AIT success Relevant to Australian practice where AIT is PBS-subsidised for specific allergens. Results suggest current clinical assessment remains optimal for patient selection rather than expensive biomarker testing Patients with grass and/or birch pollen allergic rhinitis considering allergen immunotherapy

Abstract

There are large variations in how individual patients respond to allergen immunotherapy (AIT) against grass and/or birch allergy. There are currently no reliable biomarkers to predict which patients are likely to benefit from the treatment. The purpose of this study was to examine the potential of cytokine and soluble immunologic checkpoint molecule (ICM) levels as biomarkers for AIT success. Blood samples collected before starting AIT were analyzed for concentration of 92 different cytokines and 14 different ICMs. Traditional univariable statistical analysis was performed to evaluate differences between responders and non-responders. Furthermore, both unsupervised and supervised machine learning algorithms were used for multivariable analysis of differences between the responders and non-responders and to try to identify clusters within the subjects which could potentially be linked to endotypes of allergic rhinitis. Neither univariable nor multivariable analysis showed any significant correlations between AIT outcome and pre-treatment levels of cytokines or soluble ICM. In the cluster analysis, 4 clustering algorithms consistently grouped 48 of the 60 subjects into 3 distinct clusters. However, these clusters did not correlate with clinical characteristics, indicating that the clusters are unlikely to represent actual biological endotypes. The findings of this study did not provide evidence supporting the use of pre-treatment levels of cytokines and ICM as biomarkers for AIT outcome.