Plasma protein profiles as markers of asthma severity and exposure to oral corticosteroids in U-BIOPRED and BIOAIR

Background: There is a need for non-invasive biomarkers of asthma phenotypes. Methods: 177 proteins with potential involvement in airway or systemic inflammation were measured in adult mild-to-moderate (MMA) and severe asthmatics (SA) from two European multicentre studies: a discovery cohort with baseline samples from the U-BIOPRED study (76 MMA, 358 SA) and a validation cohort from the BIOAIR study (48 MMA, 58 SA). BIOAIR included a placebo-controlled oral steroid intervention trial. Proteins were measured by a multiplex antibody bead assay and steroid use confirmed in urine by mass spectrometry. Results: Differential levels were found for 110 proteins in U-BIOPRED, the majority being elevated in SA compared to MMA. There were no differences in plasma protein profiles between MMA and healthy controls. Proteins confirmed as elevated in SA in both U-BIOPRED and BIOAIR were alpha-1-antichymotrypsin, apolipoprotein-E, CCL23, complement proteins C9 and factor I, glutathione S-transferase, IL-6, sphingomyelin phosphodiesterase 3, RANK and TGF-β1. The controlled steroid intervention in BIOAIR influenced the plasma levels of 144 proteins and the majority decreased. In U-BIOPRED, some of the findings from the steroid trial were supported by comparing levels between SA taking and not taking oral steroids. In analysis limited to asthmatics without any steroid use, most of the associations with severity remained. Conclusions: Proteins in plasma were associated with asthma, and severity was reflected by elevated levels. Some proteins were affected by both severity and steroid use, highlighting the importance to take pharmacological treatment into consideration in biomarker studies.