Rationale: Little is known about the use of biomarkers in guiding treatment decisions in routine asthma management. The objective of this study was to determine whether adding a LABA to an ICS would control bronchial hyperresponsiveness (BHR) at an overall lower dose of ICS when titration of medication was based upon the assessment of routine clinical measures with or without the measurement of BHR.

Methods: After a 2-week run-in period, subjects (> or = 12 years) were randomized to one of three treatment groups. Two groups followed a BHR treatment strategy (based on clinical parameters [lung function, asthma symptoms, and bronchodilator use] and BHR) and were treated with either fluticasone propionate/salmeterol (FSC(BHR) group) or fluticasone propionate (FP(BHR) group) (n=156 each). The third group followed a clinical treatment algorithm (based on clinical parameters alone) and were treated with fluticasone propionate (FP(REF) group; n=154). All treatments were administered via Diskus. Treatment doses were adjusted as needed every 8 weeks for 40 weeks according to the subject's derived severity class, which was based on clinical measures of asthma control with or without BHR.

Results: The mean total daily inhaled corticosteroids (ICS) dose during the double-blind treatment period was lower, although not statistically significant, in the FSC(BHR) group compared with the FP(BHR) group (a difference of -42.9 mcg; p=0.07). Compared with the FP(REF) group, the mean total daily ICS dose was higher in the FSC(BHR) group (a difference of 85.2 mcg) and was significantly higher in the FP(BHR) group (a difference of 131.2 mcg, p=0.037).

Conclusion: This study demonstrated that for most subjects, control of BHR was maintained when treatment was directed toward control of clinical parameters. In addition, there was a trend towards control of BHR and clinical measures at a lower dose of ICS when used concurrently with salmeterol.