Phenotypic predictors of long-term response to inhaled corticosteroid and leukotriene modifier therapies in pediatric asthma.

BACKGROUND In children with mild-to-moderate persistent asthma, identification of phenotypic predictors to guide selection of a controller regimen is essential. OBJECTIVE We sought to identify phenotypic characteristics having predictive value for the difference in treatment responses between twice-daily fluticasone and once-daily montelukast. METHODS Data from the Pediatric Asthma Controller Trial were assessed with multivariate analysis. Outcomes included the change in asthma control days (ACDs), FEV(1), peak expiratory flow, and time to first asthma exacerbation measured over a 1-year treatment period. RESULTS The mean age was 9.6 +/- 2.1 years, 60% were male, 50% had a parental history of asthma, and 78% had positive aeroallergen skin prick test responses. The mean percent predicted prebronchodilator FEV(1) was 97.8% +/- 12.9%, the median PC(20) value was 0.93 mg/mL, and the median exhaled nitric oxide (eNO) level was 25.2 ppb. A history of parental asthma best predicted the expected treatment benefit with fluticasone compared with montelukast in terms of gain in ACDs (adjusted P = .02) and time to first exacerbation (adjusted P = .05). Increased baseline eNO levels predicted the differential treatment response for fluticasone regarding the gain in ACDs (adjusted P = .01). Prior inhaled corticosteroid (ICS) use (adjusted P = .01) and low PC(20) values (adjusted P = .03) each predicted the expected treatment benefit with fluticasone over montelukast regarding time to first exacerbation. No phenotypic characteristics predicted treatment benefits for montelukast over fluticasone for either outcome. CONCLUSIONS Physicians treating children with a parental history of asthma, increased eNO levels, low PC(20) values, or a history of ICS use can expect the best long-term outcomes with ICS therapy compared with treatment with leukotriene receptor antagonists.