D2.158 - Circulating 1-methylnicotinamide can predict the treatment response of dupilumab in adult asthma

Severe eosinophilic asthma is increasingly managed with biologics, but these therapies are costly and require prolonged treatment before response can be assessed. Current clinical indicators, such as blood eosinophils and fractional exhaled nitric oxide (FeNO), have limited predictive value, underscoring the need for reliable biomarkers. We aimed to identify plasma metabolite biomarkers predictive of biologic response by developing a predictive model. 

Baseline plasma samples (N=275) from individuals receiving biologics (dupilumab [n=150], mepolizumab [n=31] and reslizumab [n=42]) or conventional treatment (T2 high conventional [n=25] and T2 low [n=27]) in the Precision Medicine Intervention in Severe Asthma (PRISM) cohort were analyzed using untargeted metabolomics to identify candidate biomarkers. The selected metabolite was subsequently quantified in an expanded cohort (dupilumab, n=208) and incorporated, along with clinical indicators, into logistic regression models. Model performance was compared across variables.

1-Methylnicotinamide (1-MNA) was identified as a candidate biomarker and was significantly elevated in dupilumab responders compared with non-responders. Logistic regression models using 1-MNA alone (ROC-AUC = 0.723) or in combination with FeNO (ROC-AUC = 0.754)—a clinical marker currently considered in dupilumab treatment decisions—demonstrated superior predictive performance compared with FeNO alone (ROC-AUC = 0.649). 

Plasma 1-MNA, identified through untargeted and targeted metabolomics, is a promising biomarker for predicting response to severe asthma treatment. Predictive modeling shows 1-MNA provides stronger discrimination between responders and non-responders than conventional clinical markers. These findings suggest that 1-MNA could support biomarker-guided selection of biologic therapy, although external validation and mechanistic studies are needed to confirm its clinical utility.